Mailinglist Archive: opensuse-project (465 mails)

< Previous Next >
Re: [opensuse-project] Re: Strategy discussion @ forums
  • From: "Carlos E. R." <carlos.e.r@xxxxxxxxxxxx>
  • Date: Mon, 05 Jul 2010 13:42:29 +0200
  • Message-id: <4C31C525.9050907@xxxxxxxxxxxx>
Hash: SHA256

On 2010-06-29 22:34, Jim Henderson wrote:
On Tue, 29 Jun 2010 10:33:25 +0200, Carlos E. R. wrote:

Makes sense to me - though in the online world, I find that as front
line support goes, there tends to be very strong technical expertise
when you operate as a meritocracy. Those who give good help continue
to do so, and tend to acquire a lot of knowledge and be able to solve a
lot of problems.

In the "real world" this does not often happen, because those in that
1st line are under-payed and unmotivated, in my experience, so that they
seek a different post, or even better, a different company, going away
as soon as they can with their experience - lost. In my circle of
friends we call them "flower pots", because attempting to solve a
problem talking to them is like talking to pottery. Very nice pottery
when it is not phone, but human. Usually young girls hired for events.
Thus the term: flower pots. Not trying to demean them, I have been a
flower pot myself, of the rough kind ;-)

Oh, sure, in the real world, front-line support is typically (but not
always) low-paid, entry-level positions. But we're talking about the
online world here; that kind of escalation is quite popular, not just in
the Novell forums, but also the Microsoft forums and other online
communities. The "front line" there are the NKPs (for Novell) or the
MVPs (for Microsoft) - and many stay there for years, completely unpaid
(there are usually other perks, conferences and the like). As I
mentioned earlier, I was one of those volunteers for more than a decade
for Novell. :)

Ah, I have never seen that, personally, it is new to me.

I think it could; the developers have the same constraints whether
they're paid to develop or not, and the technical people answering the
questions (in my experience) have been volunteers. I used to be one
myself in the Novell forums years ago.

The primary difference with an open project is that of accessibility to
the "back line" and to the developers. But the mechanism of filtering
those escalations is something that I see as equally valid regardless
(not talking about 'proxy bug reports' here, just the filtering of the
"I can't print" type posts from backline/development level assistance
unless there's an actual problem that needs to be fixed).

And it would make for faster solving. It is discouraging to find that
the first answer on a bugzilla that took many hours to investigate and
report is not written till several months later. Perhaps better
"triaging" (is that the word?) would help.

Triaging would apply to coding a quick fix; it would be additional
troubleshooting before a bug is raised. The ideal workflow, I think is:

1. User reports problem (through forums, MLs, whatever) and asks if it's
been seen before
1a. User perhaps also searches bugzilla
2. User works with a 'front-line' SME to determine if it's a bug,
configuration issue, or training issue (ie, the user didn't know what
they needed to know).
3. If the problem is a bug, then the bug is reported by someone who has
seen the problem (ie, the original poster or someone who has reproduced
it) in bugzilla.
4. If the bug is a duplicate that just couldn't be found, the bug is
flagged as a duplicate; otherwise, the bug is assigned a priority and put
in the queue for a fix.

I've probably missed a few steps here, but this is the high level view of
what I see as an 'ideal' flow (largely because it's the flow I tend to
use myself).

I was rather thinking on the line of a group of people reading all new
bugzillas, doing a fast
assessment, asking for the typical logs or whatever is missing, all at most 36
hours after the
initial report. Then advise the reporter about what is the typical response
time on that kind of
problem considering the current backlog. And forward to the proper team of
people that handle that
particular type of problem, if known.

Sometimes I have reported a problem and then, 6 months later I get asked (first
answer) to add some
info or a log or whether I also had that other symptom. By that time I don't
remember! Plus it is
very discouraging for a reporter not to get a simple feedback in months.

- --
Cheers / Saludos,

Carlos E. R.
(from 11.2 x86_64 "Emerald" GM (Minas Tirith))
Version: GnuPG v2.0.12 (GNU/Linux)
Comment: Using GnuPG with SUSE -

To unsubscribe, e-mail: opensuse-project+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse-project+help@xxxxxxxxxxxx

< Previous Next >
Follow Ups