Adding rules 0 and 0.1 ?
Pierre Poulain
mailing lists and forums" project,
I have been informed via Tweeter of your project that, I believe, is
both interesting and needed.
Before any modification on the Wiki gene page, I would like to submit
you two "new" rules.
Maybe the very first rule (number 0) could be "Define your problem" or
"What's your problem?".
I think it is really important to clearly identify where is the
problem. Is this a software issue?
Is it a data (quality? quantity?) issue? Is it an algorithm/method
problem? (e.g. How to compute a
RMSD between two proteins?) Is it a more wider science problem? (e.g
Where could I find
structures of proteins?) and most important of all, Is your problem
reproducible? And in which conditions?
This "0" rule is strongly related to actual rules 4 and 5, but could
(should), in my opinion, be more emphasized
The immediate following rule to the latter could be "Search deep and
wide". This means that
before asking / posting any question, it is really important to widely
browse the internet (as a whole not
only forums and mailing lists -- blogs could be really informative
too).
Indeed, what is the point to post a question in a mailing list if the
issue has already been tackled
by a researcher on her/his blog or even on Tweeter?
This so-called "0.1" rule is of course a generalization of actual rule
3.
I hope this help in the discussion.
Best regards,
Pierre
--
Dr. Pierre Poulain
DSIMB team
Inserm U665 and Univ. Paris Diderot-Paris 7, France
http://www.dsimb.inserm.fr/~poulain/
Colin Gillespie
> I have been informed via Tweeter of your project that, I believe, is
> both interesting and needed.
Ditto.
>
> Before any modification on the Wiki gene page, I would like to submit
> you two "new" rules.
Just to add my 2p worth.
>
> Maybe the very first rule (number 0) could be "Define your problem" or
> "What's your problem?".
It should emphasise that if you don't know what the problem is, then
no-one will be able to help you with it.
> I think it is really important to clearly identify where is the
> problem. Is this a software issue?
> Is it a data (quality? quantity?) issue? Is it an algorithm/method
> problem? (e.g. How to compute a
> RMSD between two proteins?) Is it a more wider science problem? (e.g
> Where could I find
> structures of proteins?) and most important of all, Is your problem
> reproducible? And in which conditions?
If your question has many complex parts, can you separate it into
multiple questions.
> This "0" rule is strongly related to actual rules 4 and 5, but could
> (should), in my opinion, be more emphasized
A nice quote from the R mailing list could be slotted in here:
<begin quote>
Just now I had an apparently insurmountable problem that's been
bugging me for days, but phrasing my question in a form suitable for
the R-help list enabled me to solve my own problem in two minutes
flat.<x>Thanks everyone.
Robin Hankin, R-help, March 2005
</end quote>
> The immediate following rule to the latter could be "Search deep and
> wide". This means that
> before asking / posting any question, it is really important to widely
> browse the internet (as a whole not
> only forums and mailing lists -- blogs could be really informative
> too).
> Indeed, what is the point to post a question in a mailing list if the
> issue has already been tackled
> by a researcher on her/his blog or even on Tweeter?
> This so-called "0.1" rule is of course a generalization of actual rule
> 3.
To add to this, if you find something that is similar, but doesn't
quite work, then provide the link. It could potentially help the
people who answer your question.
Cheers
Colin
--
Dr Colin Gillespie
http://www.mas.ncl.ac.uk/~ncsg3/
Giovanni Marco Dall'Olio
Hi Giovanni and others,
Ditto.
> I have been informed via Tweeter of your project that, I believe, is
> both interesting and needed.
thank you for joining this initiative. I hope that inviting many people will lead to a better and more useful document.
> Maybe the very first rule (number 0) could be "Define your problem" or
> "What's your problem?".
It should emphasise that if you don't know what the problem is, then
no-one will be able to help you with it.
Good idea, we definitely should have a rule to say that it is important to define well the problem.
A good question should be such that any person who knows the solution should be able to answer in a single message, without having to ask for clarification and without hidden details .
StackEngine Websites are built following this logic: if you look at, for example, http://biostar.stackexchange.com/ , you will see that participants are allowed to submit only one answer per question, and there is only one level of questioning/answering.
To include this rule, we need to provide some guidelines, and examples of well and badly posed questions.
The quote from the R mailing list is great, let's see if we can find other ones.
If your question has many complex parts, can you separate it into
> I think it is really important to clearly identify where is the
> problem. Is this a software issue?
> Is it a data (quality? quantity?) issue? Is it an algorithm/method
> problem? (e.g. How to compute a
> RMSD between two proteins?) Is it a more wider science problem? (e.g
> Where could I find
> structures of proteins?) and most important of all, Is your problem
> reproducible? And in which conditions?
multiple questions.
Good point. Do you think that this should go as a separate rule, "Ask only a single question per discussion", or should be included in the previous one?
Let's see how much space we will have for each rule.
To add to this, if you find something that is similar, but doesn't
> The immediate following rule to the latter could be "Search deep and
> wide". This means that
> before asking / posting any question, it is really important to widely
> browse the internet (as a whole not
> only forums and mailing lists -- blogs could be really informative
> too).
> Indeed, what is the point to post a question in a mailing list if the
> issue has already been tackled
> by a researcher on her/his blog or even on Tweeter?
> This so-called "0.1" rule is of course a generalization of actual rule
> 3.
quite work, then provide the link. It could potentially help the
people who answer your question.
These are also good ideas. Rule 3 says to not ask a question that has already been asked in the same place; we could add these two concepts in the text of rule 3.
Depending on where the issue has been answered, it could be acceptable to ask it even if we already know that is has already been solved elsewhere, as other people may point out to different solutions. And leaving a link to documents where the issue has already been tracked is useful, both for the people who read the question (who will find an useful blog), than for the author of the blog, who sees his/her pagerank grow.
Cheers
Colin
--
Dr Colin Gillespie
http://www.mas.ncl.ac.uk/~ncsg3/
--
Giovanni Dall'Olio, phd student
Department of Biologia Evolutiva at CEXS-UPF (Barcelona, Spain)
My blog on bioinformatics: http://bioinfoblog.it
Pierre Poulain
> A good question should be such that any person who knows the solution should
> be able to answer in a single message, without having to ask for
> clarification and without hidden details .
> To include this rule, we need to provide some guidelines, and examples of
> well and badly posed questions.
> The quote from the R mailing list is great, let's see if we can find other
> ones.
manuscript.
However, I believe that to keep the paper reasonably short (as all
articles in the
"Ten simple rules...") we should avoid quoting too many times.
> Good point. Do you think that this should go as a separate rule, "Ask only a
> single question per discussion", or should be included in the previous one?
> Let's see how much space we will have for each rule.
"Define your problem simply and unambiguously"
> Depending on where the issue has been answered, it could be acceptable to
> ask it even if we already know that is has already been solved elsewhere, as
> other people may point out to different solutions. And leaving a link to
> documents where the issue has already been tracked is useful, both for the
> people who read the question (who will find an useful blog), than for the
> author of the blog, who sees his/her pagerank grow.
- Link to similar things
- Link to things that do not work
- Link to previous answer / solution
- Link to external resources (blog, website and other mailing list)
Given the growth of available knowledge available on the web, the
issue in
a few years won't be " ten rules for contributing to mailing lists and
forums"
but "ten rules to find the answer to your (already solved) problem in
the big internet" :-)
Cheers
Pierre