poorejc commented on issue #42:
URL: 
https://github.com/apache/incubator-flagon-useralejs/issues/42#issuecomment-796378855


   Comments inline!
   
   
   > Greetings:
   > We would like to get the following basic information:
   
   Lots of this info isnt basic--people pay people or services for some of this 
stuff. I can give what we have, and we can help build what you need, but we 
need tickets and could use contributions or co-development.
   
   > 
   > 1. How many unique users have used the application in a day?
   
   This is already baked into our dashboard. You can do unique users, hosts, 
whatever you want.
   
   > 2. How many unique users have used the application in a week?
   
   This is already baked into our dashboard. Just a time filter on the same 
stuff as above.
   
   > 3. How many unique users have used the application in a month?
   
   This is already baked into our dashboard. Just a time filter on the same 
stuff as above.
   
   > 4. What is the sequence of clicks taken usually by users and how many 
users use the similar sequence?
   
   This is really advanced stuff and requires some processing--same for flagon 
as for any other tool or service. This takes some segmentation, some modeling, 
and some inference. This is on my list of things to get going in distill and I 
have an idea for how to do this, but we could use some help on this. There's 
also the issue of deployment. Kibana can't do the processing, so you could work 
the analytical side as a python microservice pushing back to the same or 
different index in ES (then can be displayed in Kibana). I'd like to explore 
superset for this pipeline, though.
   
   So, no this isn't currently available.
   
   > 5. Not sure if this is possible “What columns do people normally hide to 
accomplish their work”?
   
   If you know the DOM path to that feature, or what that behavior looks like, 
then this can be listened for.
   
   > 6. How much time does a user spend on the application on an average per 
day?
   
   Tricky in Kibana, b/c it doesn't really do time-date math, but there might 
be a way to do this--would need to experiment.
   
   > 
   > Would greatly appreciate any assistance in setting up the above reports.
   
   I have a little bit of time off late next week (Thurs). I could give you an 
hour between 0900-1530 EST US. send me some times to [email protected] and I 
can set up a zoom call to give you some pointers.
   > 
   > I know you had mentioned that the company that started the project have 
changed their thought process in making advanced analytics available to the 
general audience. Any thoughts on when this would be available?
   
   Never. Sorry. I sent you all the public reports that there are on that 
earlier work. You're welcome to reverse engineer. Should be OK as long as you 
don't yourself try and sell that (some of it is patented).
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to