Re: Surveying users on use of help/web #WebHelpUpdate


JerryG
 

The results are in.  Almost 160 usable surveys (not counting developers who did the pre-test) were filed over a two week period ending 7/23 from over a dozen countries, (60%+ US, from 34 states).  I put my results document into file folder JerryG in this forum, and will distribute a similar version (without my intermediate work sheets) to the Users Group.  Please scan the almost 100 comments that people wrote in (several key ones summarized below).  Post to thread with your comments.

 

Here are a few highlights that I picked out:

90% of respondents have access to internet when working with JMRI.

 

76% use Help from a JMRI window at least some of the time.

81% access the web site for help at least some of the time.

62% ask questions in the user forum (which is interesting since you would likely have to be a user of the forum to even know about the survey - lots of lurkers).

Only 30% ask a friend for help!

 

90% find the help they need via Help/web at least some of the time but only 37% say “most of the time.”

33% find errors in help/web some of the time.

62% use the Table of Contents.

63% use the Index.

Only 34% use the Glossary.

 

29% are on 4.19x.

32% are on 4.20 (note that this was only 1-2 weeks after 4.20 came out).

Only 30% report updating to production release soon after it comes out.

37% report usually installing test releases. 

 

56% will report a problem to the user group at least some of the time

18% report a problem to the Issues List in Github

 

Collectively, they reported using over 20 other forums.

 

Some key write-in responses (edited, grouped, summarized):

Great software, appreciate all your work, "best community maintained s/w I have seen"  (more than a dozen kudos).

Have difficulty finding content, needs more cross references, etc.

Explanations not sufficiently geared to beginners (should be “like we were 5th graders”).

Need “getting started,” step-by-step, how the pieces fit together

Indicate what version the screen shots are from and that the text refers to (much is out of date or inconsistent).

Nice to have many ways to do essentially the same thing - but need comparison of pros/cons of each (e.g Logix vs scripts vs Routes, LRoutes vs Routes).

 

For the user group: Those who know nothing about a problem should refrain from saying or requesting info as if they do.

 

So I would say this was affirming of the value of help info, although pointing out the interest in simpler explanations and more “integrating” step-by-step pages (like the one Dave Sands just did for PanelPro).  The idea of labeling each page or picture with what release it refers to might help to avoid confusion as pages get out of date (while not exactly a solution to this issue,  we have just added a line to the standard Footer that provides a pointer to the page history).  So definitely some things to think about as we create/update help info.

 

I’ll post a version of this to the Users’ Forum in a couple of days.

 

Jerry

 

P.S.  Is there anything useful we can do with the names/emails of the people who provided them?  About 60.

Join jmri@jmri-developers.groups.io to automatically receive all group messages.