Is there any analytics data for the current site such as google analytics? Having this data will help prioritize the features that get developed. It's always very difficult to rebuild every feature in the MVP when replacing a website. The MVP should have only the most important ie most used features. Too often people assume what the most important features are when the analytics show otherwise. Some features are obvious (download tor) but priorities can get less clear quickly. The analytics make it much easier to focus on the right things for MVP. Sent from my iPhone
On Wed, Jan 08, 2014 at 09:21:21PM -0500, seanmrafferty@me.com wrote 0.8K bytes in 0 lines about: : Is there any analytics data for the current site such as google analytics? Having this data will help prioritize the features that get developed. It's always very difficult to rebuild every feature in the MVP when replacing a website. The MVP should have only the most important ie most used features. Too often people assume what the most important features are when the analytics show otherwise. Some features are obvious (download tor) but priorities can get less clear quickly. The analytics make it much easier to focus on the right things for MVP. There's none. We don't much in order to avoid legal attacks on users. We have log files from the webservers, but they look like this: 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /images/onion.jpg HTTP/1.1" 200 3911 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /js/jquery.min.js HTTP/1.1" 200 31940 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /dist/torbrowser/3.5/torbrowser-install-3.5_en-US.exe HTTP/1.1" 206 5411707 "https://www.torproject.org/projects/torbrowser.html.en" "-" I'm willing to just publish the raw log files, but others get all antsy about this. -- Andrew http://tpo.is/contact pgp 0x6B4D6475
I guess we could write a script to generate page counts from a log file. That would be fun and give a rough idea of what should be prioritized. There may already be something that can do this. What kind of web servers are you running? I’m not an analytics expert, but I play one on TV. Actually, I literally sit next to one at work. I will ask him about this, but would you ever consider adding some kind of analytics to the new site if it could be anonymized? Although we would love to know who uses the site, we really just need to know what they are doing. It would really help you to understand how the site is used, what works, what doesn’t work, etc. It’s much easier to make decisions and set priorities about what to work on when you have this kind of data. It’s might be clear to some what goes in the MVP, but after that you run the risk of spending limited resources on things that don’t matter as much. On Jan 8, 2014, at 10:29 PM, andrew@torproject.is wrote:
On Wed, Jan 08, 2014 at 09:21:21PM -0500, seanmrafferty@me.com wrote 0.8K bytes in 0 lines about: : Is there any analytics data for the current site such as google analytics? Having this data will help prioritize the features that get developed. It's always very difficult to rebuild every feature in the MVP when replacing a website. The MVP should have only the most important ie most used features. Too often people assume what the most important features are when the analytics show otherwise. Some features are obvious (download tor) but priorities can get less clear quickly. The analytics make it much easier to focus on the right things for MVP.
There's none. We don't much in order to avoid legal attacks on users. We have log files from the webservers, but they look like this:
0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /images/onion.jpg HTTP/1.1" 200 3911 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /js/jquery.min.js HTTP/1.1" 200 31940 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /dist/torbrowser/3.5/torbrowser-install-3.5_en-US.exe HTTP/1.1" 206 5411707 "https://www.torproject.org/projects/torbrowser.html.en" "-"
I'm willing to just publish the raw log files, but others get all antsy about this.
-- Andrew http://tpo.is/contact pgp 0x6B4D6475 ________________________________________________________________________ Tor Website Team coordination mailing-list
To unsubscribe or change other options, please visit: https://lists.torproject.org/cgi-bin/mailman/listinfo/www-team
I guess we could write a script to generate page counts from a log file. That would be fun and give a rough idea of what should be prioritized. There may
Analog (http://www.analog.cx/), Webalizer (http://www.webalizer.org/) and Awstats (http://awstats.sourceforge.net/) can generate reports from raw log files. Check 'em out :D
I’m not an analytics expert, but I play one on TV. Actually, I literally sit next to one at work. I will ask him about this, but would you ever consider adding some kind of analytics to the new site if it could be anonymized? Although we would love to know who uses the site, we really just need to know what they are doing. It would really help you to understand how the site is used, what works, what doesn’t work, etc. It’s much easier to make decisions and set priorities about what to work on when you have this kind of data. It’s might be clear to some what goes in the MVP, but after that you run the risk of spending limited resources on things that don’t matter as much.
On Jan 8, 2014, at 10:29 PM, andrew@torproject.is wrote:
On Wed, Jan 08, 2014 at 09:21:21PM -0500, seanmrafferty@me.com wrote 0.8K bytes in 0 lines about: : Is there any analytics data for the current site such as google analytics? Having this data will help prioritize the features that get developed. It's always very difficult to rebuild every feature in the MVP when replacing a website. The MVP should have only the most important ie most used features. Too often people assume what the most important features are when the analytics show otherwise. Some features are obvious (download tor) but priorities can get less clear quickly. The analytics make it much easier to focus on the right things for MVP.
There's none. We don't much in order to avoid legal attacks on users. We have log files from the webservers, but they look like this:
0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /images/onion.jpg HTTP/1.1" 200 3911 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /js/jquery.min.js HTTP/1.1" 200 31940 "https://www.torproject.org/download/download.html.en" "-" 0.0.0.1 - - [09/Jan/2014:00:00:00 +0000] "GET /dist/torbrowser/3.5/torbrowser-install-3.5_en-US.exe HTTP/1.1" 206 5411707 "https://www.torproject.org/projects/torbrowser.html.en" "-"
I'm willing to just publish the raw log files, but others get all antsy about this.
-- Andrew http://tpo.is/contact pgp 0x6B4D6475 ________________________________________________________________________ Tor Website Team coordination mailing-list
To unsubscribe or change other options, please visit: https://lists.torproject.org/cgi-bin/mailman/listinfo/www-team
Tor Website Team coordination mailing-list
To unsubscribe or change other options, please visit: https://lists.torproject.org/cgi-bin/mailman/listinfo/www-team
On Thu, Jan 09, 2014 at 10:50:29AM +0100, simeunf@gmail.com wrote 3.2K bytes in 0 lines about: : >I guess we could write a script to generate page counts from a log file. That would be fun and give a rough idea of what should be prioritized. There may : : Analog (http://www.analog.cx/), Webalizer : (http://www.webalizer.org/) and Awstats : (http://awstats.sourceforge.net/) can generate reports from raw log : files. Check 'em out :D They don't work on our log files because we strip most of the information. -- Andrew http://tpo.is/contact pgp 0x6B4D6475
On Wed, Jan 08, 2014 at 11:05:53PM -0500, seanmrafferty@me.com wrote 2.9K bytes in 0 lines about: : I guess we could write a script to generate page counts from a log file. That would be fun and give a rough idea of what should be prioritized. There may already be something that can do this. What kind of web servers are you running? I've been using scripts and 'visitors' to get some stats and pretty graphs of usage paths through the sites. One such pretty graph is https://trac.torproject.org/projects/tor/attachment/ticket/10329/2013-11-www... and another is https://trac.torproject.org/projects/tor/attachment/ticket/10329/2013-12-www.... The active ticket is https://trac.torproject.org/projects/tor/ticket/10329 visitors is http://www.hping.org/visitors/ : I’m not an analytics expert, but I play one on TV. Actually, I literally sit next to one at work. I will ask him about this, but would you ever consider adding some kind of analytics to the new site if it could be anonymized? Although we would love to know who uses the site, we really just need to know what they are doing. It would really help you to understand how the site is used, what works, what doesn’t work, etc. It’s much easier to make decisions and set priorities about what to work on when you have this kind of data. It’s might be clear to some what goes in the MVP, but after that you run the risk of spending limited resources on things that don’t matter as much. The goal of the data collection is to collect the minimum possible. Assume someone cracks the servers and leaks the log files. In a perfect world. we don't log anything. We don't want any identifying info stored in the logs. -- Andrew http://tpo.is/contact pgp 0x6B4D6475
participants (3)
-
andrew＠torproject.is -
Sean Rafferty -
Simeun Furtula