David,
We do an apache authentication for the tool, so every user has to login.
Could this be used somehow to write a value to the cookie upon login so that
all users are known?
I don't care about the metrics being very accurate. I just want to know who
spent time looking at the diffs. Right now, we get the situation where the
majority (maybe all?) the time is logged against the unknown user, so there
is no way to know if anyone really looked at the diffs.
Thanks,
David
On Mon, Jun 9, 2008 at 12:38 AM, David Sitsky <david.sitsky@...>
wrote:
> Hi David,
>
> Really the facility that is there now is only to give rough
> indications of time spent without introducing the notion of a
> formalised user into the system. It relies on cookies, and as you
> have stated, is far from perfect.
>
> Even with proper user authentication, it would still a rough estimate
> of time spent. A user could open a page, get a cup of coffee, go to
> lunch, then add a few comments, and the system really has no accurate
> way of knowing how long they really spent looking at the review.
>
> For better statistics, it is actually better to get the users to
> actually enter in the time spent, rather than relying on the system
> itself., which is at best only very rough estimates. This is what
> teams do that want higher-quality metrics - but this does require more
> dicipline from the reviewers, which may or may not work depending on
> the team.
>
> Cheers,
> David
>
> On Thu, Jun 5, 2008 at 11:27 PM, David Carson <dccarson@...> wrote:
> > Let my get clarification on two scenarios:
> > 1. User is one of the defined reviewers. Is it possible for this user to
> > show up as the unknown user? Maybe they have cookies turned off, maybe
> they
> > never submit a comment but do spend lots of time looking at diffs.
> > 2. User is not listed in reviewers or cc list. Is this intruder more or
> > less likely to show up as the unknown user?
> >
> > Is there any way to prevent this behavior, besides just telling
> developers
> > to be careful? (That doesn't work, by the way.) This is extremely
> > undesirable from our point of view. We always want to know who spent how
> > much time, particularly if scenario #1 is possible.
> >
> > Thanks,
> > David
> >
> > On Thu, Jun 5, 2008 at 5:21 AM, David Sitsky <david.sitsky@...>
> wrote:
> >>
> >> Hi David,
> >>
> >> It is to do with cookies. Basically, the first time you make a
> >> comment or submit a review, Codestriker will remember your email
> >> address in a cookie. From that point on, the email address is used
> >> when recording actions/metrics. For users who haven't done so, or
> >> have disabled cookies, then you will get the "unknown user".
> >>
> >> Cheers,
> >> David
> >>
> >> On Thu, Jun 5, 2008 at 8:59 AM, David Carson <dccarson@...>
> wrote:
> >> > Our Codestriker 1.9.4 deployment seems to frequently attribute time
> >> > spent to
> >> > the infamous unknown user (i.e., Topic Information.User Metrics).
> This,
> >> > of
> >> > course, is of little use.
> >> >
> >> > Could you explain how this happens and what we might do to prevent it?
> >> > (I'm
> >> > wondering if it has to do with cookies, based on what little I
> >> >
> >> > could find in old discussions.)
> >> >
> >> > Thanks,
> >> > David
> >> >
> >> >
> >> >
> -------------------------------------------------------------------------
> >> > Check out the new SourceForge.net Marketplace.
> >> > It's the best place to buy or sell services for
> >> > just about anything Open Source.
> >> > http://sourceforge.net/services/buy/index.php
> >> > _______________________________________________
> >> > Codestriker-user mailing list
> >> > Codestriker-user@...
> >> > https://lists.sourceforge.net/lists/listinfo/codestriker-user
> >> >
> >> >
> >
> >
>
|