On 24/10/16 14:48, A. Soroka wrote:
I'm a bit confused by some of the last few messages in this thread. I haven't
heard anyone suggesting that anyone be excluded from anything.
The title of this thread is "raising the bar".
The root post says:
"""
* inform the subscribers of an upcoming enforcement of the publishing
rules (relevance, clearness, completeness, etc.)
* stop answering the vague/off-topic/badly presented questions
* if they insist, remind them the topic of the list and good practices
in problem reporting, and warn them of a possible ban.
"""
I take the first bullet to mean some "enforcement" is proposed and the
last bullet to suggest that a "ban" is "possible".
Dave and Rob are right to point out that projects go through phases. What's
more, the kind of communication appropriate to one phase may not be as useful
at another. I have no reason to suppose that Jena's community is growing wildly
(great though that would be!) but I'm confident that it is growing. From my
experience with open source work, the kinds of communication that are useful
for a project that involves (say) a few researchers at one or two universities
is not appropriate for a project in production at hundreds of sites, and all
along the spectrum in between styles and means of communication need to adapt.
I was not pointing out that projects go through phases.
I was pointing out that Jena has been in mature use, including by
students on course projects, for many years. We get phases where we get
naive and poorly asked questions from students. Those phases are more
related to course lifecycles than to Jena lifecycles.
Patient responses, as have been given here, generally work. If they
don't then continued such poor questions can simply go unanswered. I
really don't think there's enough volume of such traffic here as to be a
problem.
Dave
That's what I think is being pointed out here: far from pushing people away, we
need to be developing new ways to be welcoming. In that wise, Stian's ideas are
really on-point, I think.
I have been involved with a few communities that have set up (with varying success) "beginner
in the community" email lists. Those are lists wherein it is explicitly understood that no
question is bad and no assumptions should be made about questioner's backgrounds. It's a good way
to "factor out work" and a way to encourage folks who may not feel that they have enough
deep knowledge to answer difficult questions to still participate answering simpler questions and
thereby grow their own skills. That might or might not be appropriate for Jena, but it's meant as
an example of communications that might only become appropriate as a community grows.
---
A. Soroka
The University of Virginia Library
On Oct 24, 2016, at 5:29 AM, Rob Vesse <[email protected]> wrote:
Totally agree with Dave here
We always get phases of this kind of traffic. And whatever project you are
talking about there will always be users who try to run before they can walk.
This problem is not unique to our community, nor to the Apache foundation as a
whole. You will encounter this whether you are running a small open source
project or maintaining a large commercial project.
Yes we can probably be a little more robust in reminding people about how to
ask the questions. Note that we already have guidelines on asking good
questions on our website:
http://jena.apache.org/help_and_support/index.html
Some people are never going to read these, some people will never read the
manual either however much you refer them to it. That is unfortunately life.
This is supposed to be an inclusive community, we shouldn’t be excluding people
just because they don’t have years of experience of a software engineer and the
concept of MVCE drilled into them yet such that they know how to ask a good
question first time every time. Yes we should encourage them to seek help in
more appropriate ways and direct them to resources that help them to ask better
questions. But we shouldn’t turn them away because they don’t meet some ideal
about the level of discourse.
When you were a student would you have asked any better questions?
Please bear in mind that many of us benefited from western education systems
with relatively small class sizes and good access to one-to-one
tuition/assistance. Often these kinds of questions are coming from students in
education systems with huge class sizes and very little access to one-to-one
tuition/assistance. Nobody learns to ask good questions without first asking
bad questions, and some people take longer than others to get how to ask good
questions.
No one is forcing any others to read through these kinds of questions yet many
others continue to do so because we care about this project and encouraging the
community around it. People are free to filter and consume this list as they
want, if there are particular uses whose questions frustrates you no one is
stopping you from filtering those emails in your chosen email client.
Rob
On 24/10/2016 09:10, "Dave Reynolds" <[email protected]> wrote:
There have been student questions on the jena users list since long
before Jena moved into Apache. While there's a burst at the moment I
don't see it as particularly worse than historic levels.
In the past sometimes teachers have stepped in and stopped it, sometimes
not.
I'm not convinced that the number of such questions at the moment is so
great as to be a problem for other subscribers.
It is fine for the list to simply ignore the really poor/off-topic
questions, especially from serial offenders. I would not be keen any
more active steps such as "publishing rules" unless the problem were
more acute than it seems to be.
Dave
On 23/10/16 23:06, Stian Soiland-Reyes wrote:
OK, I think that is a good idea to get in touch with the teachers; perhaps
so they can give us an advance notice and we can understand what their
course is meant to teach. So a more friendly request for the teachers to
get in touch (or we ask directly the name/email of their teacher), but
without the "so you stop irritating us" bit :-). Presumably the teachers
dont want us to do the assignment for their students!
There could even be opportunities to do like a webinar or video with a
short Jena intro, there are is probably some material from Elixir's Bring
Your Own Data training events and similar that we could link to; if the
teachers have better background materials and tutorials it can hopefully
reduce our email load.
On 23 Oct 2016 10:43 pm, "A. Soroka" <[email protected]> wrote:
Then there are the obvious school examples, which seem to ask us the
actual assignment rather than Jena questions. It is fair for us to dodge
those, but perhaps in a less hostile way.
It seems to me that this is the entire question: there aren't really the
kinds of problems Colin Maudry raised _except_ with these examples. And the
messages that worry me are not the initial questions that amount to "please
do my assignment" but the fact that helpful voices on the list give in
response to such questions good advice and next steps which are repeatedly
ignored.
I think we are friendly (perhaps sometimes too helpful!), but I wouldn't
go to a "go away and talk to your teacher" route, but rather in general
respond with what is expected of a good question and what the poster should
try first.
I'm not sure if this particular remark is in response to my suggestion,
but just in case, I will clarify: I don't want to tell the students to go
away, I want to tell them to ask their teacher(s) to contact Jena directly
(instead of inadvertently and indirectly by giving assignments that show up
immediately as questions on the user list), hopefully to help create a more
appropriate kind of engagement for their students with the Jena community.
---
A. Soroka
The University of Virginia Library
On Oct 23, 2016, at 5:24 PM, Stian Soiland-Reyes <[email protected]>
wrote:
Agree to not go too aggressive in general, it could also strike down
users
who like Jena as a tool (remember we have command lines and servers!) or
have been recommended Jena, but who have not before used Java as
programming language before. Here, tutorials and examples is what we
should
point to.
Then there are the obvious school examples, which seem to ask us the
actual
assignment rather than Jena questions. It is fair for us to dodge those,
but perhaps in a less hostile way. Many students and researchers I have
interviewed in the Big Data community say they struggle to post their
questions on mailing lists for the tools they use, as they get hammered
down for basically not being geeky enough. Consequently they don't come
back when their skill sets have improved and they could potentially have
contributed back.
Also remember that students have perhaps never before used a public
mailing
list and already struggle to separate what is RDF, what is OWL, what is
Java, what is Jena, what is just a bug in their own code.
I think we are friendly (perhaps sometimes too helpful!), but I wouldn't
go
to a "go away and talk to your teacher" route, but rather in general
respond with what is expected of a good question and what the poster
should
try first. Point to gist.github.com or similar as a way to paste code
rather than getting it in the abstract ("I tried setting the literal")
helps a lot.
Also I think we can reply shorter (but friendly) as a bounce, rather
than a
complete reply to help them with the more obvious assignment side. We can
point to tutorials for coding as well; Software Carpentry has many great
starting points.
On 23 Oct 2016 7:43 pm, "Paul Houle" <[email protected]> wrote:
I find this thread disturbing. Many people in the RDF community have
worked a long time and it's just recently that the uptake has broadened
(people are looking at JSON-LD and starting to understand what it means,
not what any particular authority says that it means, but what it
actually means.)
I do believe that problems should be made reproducable and as a group we
could industrialize that. For instance, a test project that can be
forked in github would be a great place to put in a query, put in a
graph, and then put in some rules at which point they could ask good
questions.
I carefully read the answers to the bad questions because I am intensely
curious about strange details in Jena that trip people up.
--
Paul Houle
[email protected]
On Sun, Oct 23, 2016, at 06:07 AM, Colin Maudry wrote:
Dear Jena developers,
Upon Andy Seaborne’s suggestion, I would like to share with you a
concern we have with certain posts shared on [email protected].
In the last couple months, we have seen certain users repeatedly
sending
questions that are either:
* hardly related to Jena and Fuseki
* very basic questions about RDF or SPARQL
* betraying the lack of common knowledge in Java programming and
coding good practice in general
What’s worse, these users, in spite of repeated remarks, keep on being
very vague in their questions, requiring the most patient subscribers
to
ask many questions just to obtain a decent understanding of the
problem.
A problem that is, again, often not much related to Jena or Fuseki.
As a subscriber, I’m tired of their consistent failure to propose clear
and concise questions and I wish the patient people who answer them
spend their mailing time on more interesting threads. I also fear it
makes certain subscribers silently go away because of this “noise”.
I first thought of publicly complaining to these users, but I thought
that the managers of the Jena lists should discuss it and take the
appropriate measures.
My suggestion is to:
* inform the subscribers of an upcoming enforcement of the publishing
rules (relevance, clearness, completeness, etc.)
* stop answering the vague/off-topic/badly presented questions
* if they insist, remind them the topic of the list and good practices
in problem reporting, and warn them of a possible ban.
Thanks for your attention,
Colin Maudry
https://twitter.com/CMaudry
