[wingide-users] Re: How about a uservoice site for Wing IDE?

Bernt Røskar Brenna bernt.brenna at gmail.com
Wed Dec 16 06:44:40 MST 2009


My request for a uservoice site got a few responses. Some comments:

Wing IDE Support wrote:
> We'll look into and possibly experiment with it or something like it; it
> does look potentially useful even to someone like me who thinks that
> email is usually the best way to communicate.
>

Email can be a nice way to communicate, I agree. But I believe that
for issues where there is a need for follow-up and discussion, e-mail
sometimes can be a bit messy.

> We always accept bug reports and feature requests that are mailed to
> support at wingware.com or to this list and try to respond quickly to each
> one.  These requests do alter our priorities, though unfortunately we
> can't do everything at once ;).

While it is great that you respond to issues mailed to you, I would
still like to have the possibility to check the status of my bug
reports and issues.

For instance, I reported a possible bug with the 'OS Commands' tool on
October 29, and got a response from support the same day, saying
"We'll try to fix them in a future release." That is the last thing I
know of this possible bug. Stuff it would be nice to know: Have you
been able to reproduce? Are others seeing the same thing? Will you try
to fix it? In what release? etc. Having this kind of information
available requires some sort of public issue/bug/feature request
tracking system.

Charles Hartman wrote:
> Two cents:  I for one don't much like it.  The "voting" model gives
> feedback in distorted ways (see the American political system for
> examples).  Wingware is so extraordinarily and exceptionally
> responsive to the one-to-one and several-to-many forum offered by
> email, intensely labor-intensive as it is for them, that I think the
> shift to a bug-tracker system would result in a much less finely tuned
> relation to the needs of users.  Users have individual needs, which

I see your point on voting - in part. But the Wing IDE development
process will not be a democratic. Since Wingware owns the source code
and all rights to it, only they decide what changes are made to the
product.

Having a public bug/issue tracker and/or feature voting system is
merely a (IMO) better way of getting user input than email.

> may coincide closely or not.  Also, Wingware obviously responds in
> part according to a series of short-range and long-range plans, and an
> understanding of what changes can be made with smaller alterations in
> the code-base and which would require big shifts in direction.  With a
> bug-voting system, either they'd be forced to override those internal
> understandings of the logical and feasible--which would tend to make
> the code incoherent and the application less robust--or they'd have to
> deal with outcry about how "63% of users want this and only 41% want
> that so why did you choose that?"

This depends on the expectations to the voting system by Wingware and
the user community. The understanding should be that the user input is
suggestions only.

Wingware support wrote:
> I see a couple of other points against a public bug tracker.  These are
> what has kept us away from adding one so far:
>
> * It takes work to run a public bug tracker (merging dups, dealing
> with spam, etc).  I'd  rather spend that time working on code and
> delivering fixes and new features.

Running a mailing list also takes time.

> * We will still get bug reports emailed privately and have a large
> existing internal database.  It's unclear how to reconcile those
> and/or sanitize private data safely without still more overhead. (And
> no, we are not going to ask customers to enter a bug when they've
> already emailed us!)

Email/issue tracking integration is pretty much a solved problem. Most
tracker solutions (JIRA, for instance) has a email interface. E-mails
sent to an address are automatically added as tickets to the system
(you then categorize and work with the issue in the tracking system
from there).

Entering an e-mailed bug can then be as easy as forwarding the e-mail
from the customer.

> * Bug trackers always have a lot of cruft in them.  See for example
> bugs.python.org.  There are 2500 open bugs, many of them well over a
> year old.  Yet, I've really only run into 2 or 3 bugs in Python myself
> in over ten years of intensive use.  The size of the bug list does not
> really reflect usability of the software, although it may be perceived
> as such if we post it publicly.  (And, believe me, our internal bug
> database is _huge_!)

Use a system that lets you have private/public issues.

Removing cruft is partly a maintenance issue, and also about updating
statuses (WontFix/UnableToReproduce/FixedInVersionxxx) and closing
issues.

> Given these points, I'm not convinced a public bug tracker or voting system
> would make us more effective in serving our customers.

You are of course in a better position to have an opinion on that.

I believe the top benefits of replacing the mailing list with a
bug/voting system would be:

1. Improve the feedback to users on issue/bugfix progress
2. Allow more users to participate, in an easier fashion. I believe
that the bar is higher for contributing to a mailing list than a
web-based system.

Just my 2 øre.


Bernt


More information about the wingide-users mailing list