Resending this. Apparently I used the wrong email address to send this from, so it didn't go to the list.
-----Original Message----- From: brynn Sent: Sunday, February 25, 2018 12:25 PM To: Martin Owens ; Ryan Gorley ; inkscape-devel@lists.sourceforge.net Subject: Re: [Inkscape-devel] Visitor analytics
What I like the best about it (Piwik/Matomo) is that it seems to only report on what I refer to as human visitors. Personally I find that much more useful. (Although I could imagine that some sites (or webmasters, owners) might be more interested in aalllll the traffic.)
So that's the sandcats.io script that I see in Firefox's No Script? So already, visitors haven't been notified about the site using this tracking?
So the login via github has to do with sandstorm, from what I can tell? Because I just log in on the page where I installed it, on my site.
brynn
-----Original Message----- From: Martin Owens Sent: Sunday, February 25, 2018 2:12 AM To: brynn ; Ryan Gorley ; inkscape-devel@lists.sourceforge.net Subject: Re: [Inkscape-devel] Visitor analytics
On Sun, 2018-02-25 at 01:24 -0700, brynn wrote:
Whatever a Piwik "sandcat sandstorm grain" is, it's probably very different from what I'm using on my site.
That's just the way it's deployed. It's easier to do that then to set it up raw although it comes with it's own oddness hence the login restrictions.
The graphs even now are interesting.
Martin,
On Tue, 2018-02-27 at 10:29 -0700, brynn wrote:
Resending this. Apparently I used the wrong email address to send this from, so it didn't go to the list.
Thanks for correcting the address. I think people have been out of the loop on this development.
I made a sandstorm based Piwik analytics service and have asked the website to put information about vistors there.
So that's the sandcats.io script that I see in Firefox's No Script? So already, visitors haven't been notified about the site using this tracking?
They're not being tracked as users, it's just for the arrigates. Although tightening the configuration will be a good thing to make sure. If we go with this solution, I plan on NOT using the django piwik app which tracks usernames and a bunch of other things.
So the login via github has to do with sandstorm, from what I can tell? Because I just log in on the page where I installed it, on my site.
Yes, this is a sandstorm thing, it replaces the login for any service you set up. A bit silly IMO, but hey ho, it was easy to get running.
Here's a picture of what we're currently getting:
Just seeing this page is so insightful. Any chance I can get access to look at these more closely?
Ryan Gorley Founder + Creative Director
https://dijt.co 1.801.999.1530 ×101 1.801.898.7926
On 02/27/2018 10:44 AM, Martin Owens wrote:
On Tue, 2018-02-27 at 10:29 -0700, brynn wrote:
Resending this. Apparently I used the wrong email address to send this from, so it didn't go to the list.
Thanks for correcting the address. I think people have been out of the loop on this development.
I made a sandstorm based Piwik analytics service and have asked the website to put information about vistors there.
So that's the sandcats.io script that I see in Firefox's No Script? So already, visitors haven't been notified about the site using this tracking?
They're not being tracked as users, it's just for the arrigates. Although tightening the configuration will be a good thing to make sure. If we go with this solution, I plan on NOT using the django piwik app which tracks usernames and a bunch of other things.
So the login via github has to do with sandstorm, from what I can tell? Because I just log in on the page where I installed it, on my site.
Yes, this is a sandstorm thing, it replaces the login for any service you set up. A bit silly IMO, but hey ho, it was easy to get running.
Here's a picture of what we're currently getting:
Ok, I looked up "arrigates" in all the places I know to learn computer terms. No luck.
DDG defaults to arrogates, which is not a very flattering claim to be making, if we're talking about not arrogating the info.
But when I forced DDG to search "arrigates", then it defaulted to aggregates. Aggregates /could/ fit in your comments.
A little help with what you're collecting?
I see the image showing your results. They look very familiar as part of Piwik/Matomo. Just general info, and no IP addresses. But I don't see how that's just the basic Piwik installation (as you said before - or I thought you did) Not unless there are just some options that I've never found before?
Oh ok, I think I see what you did. You just don't have those particular features with IP addresses enabled (the widgets). (such as Real Time Visits) So the info could be collected, if you enabled those features, right? If you opened the Visitor Log, you would see the individual visitors. You're just not going to open it, right?
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
All best, brynn
-----Original Message----- From: Martin Owens Sent: Tuesday, February 27, 2018 10:44 AM To: brynn ; Inkscape-Devel ; Ryan Gorley Subject: Re: Fw: [Inkscape-devel] Visitor analytics
On Tue, 2018-02-27 at 10:29 -0700, brynn wrote:
Resending this. Apparently I used the wrong email address to send this from, so it didn't go to the list.
Thanks for correcting the address. I think people have been out of the loop on this development.
I made a sandstorm based Piwik analytics service and have asked the website to put information about vistors there.
So that's the sandcats.io script that I see in Firefox's No Script? So already, visitors haven't been notified about the site using this tracking?
They're not being tracked as users, it's just for the arrigates. Although tightening the configuration will be a good thing to make sure. If we go with this solution, I plan on NOT using the django piwik app which tracks usernames and a bunch of other things.
So the login via github has to do with sandstorm, from what I can tell? Because I just log in on the page where I installed it, on my site.
Yes, this is a sandstorm thing, it replaces the login for any service you set up. A bit silly IMO, but hey ho, it was easy to get running.
Here's a picture of what we're currently getting:
Am 28.02.2018 um 00:52 schrieb brynn:
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree: Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
The kind of information mentioned so far can easily be collected in anonymized form and I hope we can give our visitors the guarantee no identifying information will ever be stored (or transmitted).
I'd favour a solution that does not involve third-party servers (what Martin set up seems to depend on it) - then again it makes it easier for people to block it ;-) (On that note you might also want to consider statistics usually only represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
Best Regards, Eduard
On Wed, 2018-02-28 at 01:23 +0100, Eduard Braun wrote:
Am 28.02.2018 um 00:52 schrieb brynn:
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree: Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
The kind of information mentioned so far can easily be collected in anonymized form and I hope we can give our visitors the guarantee no identifying information will ever be stored (or transmitted).
I'd favour a solution that does not involve third-party servers (what Martin set up seems to depend on it) - then again it makes it easier for people to block it ;-) (On that note you might also want to consider statistics usually only represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
No the service is ours, our machine, our databases. I think only the dns is from sandstorm
Martin,
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree:
Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
I think if the information falls into the wrong hands, then our trust was misplaced, and we did not protect the info well enough. If a group takes on such a trust, then it better be protecting the information appropriately.
Maybe there is some product which only can collect info which is not traced to the IP address. Or if not now, I suspect there will be pretty soon. But for now, I'm not sure if there is any kind of analytics which doesn't trace to the IP address.
That gives us the choice of having some analytics, and being trust-worthy and taking good care of that info (including deleting the parts we don't need). Or just not having any analytics.
At least that's how I see it. If there is some analytics app which only gives non-personally identifying info, then that's what we should use. I just don't know if there is any such program.
(On that note you might also want to consider statistics usually only
represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
Ah yes, I am aware that I'm not getting statistics on people who don't want to be tracked and are aware how to avoid it. It's interesting, when you think of the Inkscape community, you're thinking mostly of developers. But the Inkscape users we see in forums, most are not very tech savvy. Rather than not caring about privacy and security, often they don't know anything about it. (More than once or twice I've referred people to security forums.)
All best, brynn
-----Original Message----- From: Eduard Braun Sent: Tuesday, February 27, 2018 5:23 PM To: brynn ; Martin Owens ; Inkscape-Devel ; Ryan Gorley Subject: Re: [Inkscape-devel] Visitor analytics
Am 28.02.2018 um 00:52 schrieb brynn:
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree: Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
The kind of information mentioned so far can easily be collected in anonymized form and I hope we can give our visitors the guarantee no identifying information will ever be stored (or transmitted).
I'd favour a solution that does not involve third-party servers (what Martin set up seems to depend on it) - then again it makes it easier for people to block it ;-) (On that note you might also want to consider statistics usually only represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
Best Regards, Eduard
Brynn,
Can I direct your attention to a page I just found documenting the features for privacy: https://matomo.org/docs/privacy/
This should allow us to annoymise the IP Addresses as well as a fair few other things.
One interesting metric we picked out from the stats, social media (see image)
So far it seems to say that Youtube is VERY important to inkscape. Mostly to windows users. Mac users are most likely to use GitHub and Reddit while Linux users are fairly spread out except for Vkontakte (which is a russian social media site)
Some other interesting parts, wikipedia and gimp provide the greatest number of cross links (we should be providing links to gimp from our home page IMO)
Forum wise, inkscapeforum.com provides the most links, but also forums.cnetfrance.fr and forum.inkscapecommunity.com, mac- forums.com, gimpforums.com, forums.adobe.com, forums.sketchup.com even forums.autodesk.com
Best Regards, Martin Owens
On Tue, 2018-02-27 at 22:29 -0700, brynn wrote:
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree:
Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
I think if the information falls into the wrong hands, then our trust was misplaced, and we did not protect the info well enough. If a group takes on such a trust, then it better be protecting the information appropriately.
Maybe there is some product which only can collect info which is not traced to the IP address. Or if not now, I suspect there will be pretty soon. But for now, I'm not sure if there is any kind of analytics which doesn't trace to the IP address.
That gives us the choice of having some analytics, and being trust- worthy and taking good care of that info (including deleting the parts we don't need). Or just not having any analytics.
At least that's how I see it. If there is some analytics app which only gives non-personally identifying info, then that's what we should use. I just don't know if there is any such program.
> (On that note you might also want to consider statistics usually only represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
Ah yes, I am aware that I'm not getting statistics on people who don't want to be tracked and are aware how to avoid it. It's interesting, when you think of the Inkscape community, you're thinking mostly of developers. But the Inkscape users we see in forums, most are not very tech savvy. Rather than not caring about privacy and security, often they don't know anything about it. (More than once or twice I've referred people to security forums.)
All best, brynn
-----Original Message----- From: Eduard Braun Sent: Tuesday, February 27, 2018 5:23 PM To: brynn ; Martin Owens ; Inkscape-Devel ; Ryan Gorley Subject: Re: [Inkscape-devel] Visitor analytics
Am 28.02.2018 um 00:52 schrieb brynn:
Well this goes back to my original comments. Collecting the info isn't bad, by itself. It comes down to the intentions of the people collecting the info. And as long as we trust ourselves, there's no problem.
Only a quick comment as you've said that before and I *strongly* disagree: Real privacy/security is not about trust - it's about guaranteeing that whatever information we collect can not be misused, even if it falls into the wrong hands.
The kind of information mentioned so far can easily be collected in anonymized form and I hope we can give our visitors the guarantee no identifying information will ever be stored (or transmitted).
I'd favour a solution that does not involve third-party servers (what Martin set up seems to depend on it) - then again it makes it easier for people to block it ;-) (On that note you might also want to consider statistics usually only represent those people who do not care about privacy and therefore will likely not catch most of the tech-savvy folks which makes up a large group of Inkscape community).
Best Regards, Eduard
participants (4)
-
brynn -
Eduard Braun -
Martin Owens -
Ryan Gorley