I'm disappointed that no OpenLibrary or Internet Archive staff ever replied
to this thread.  I guess that shows where community communication ranks in
the priority scheme of things.

Below are some comments on the replies from Karen and Charles (thanks for
taking the time to reply!).  Perhaps reviving the thread will stimulate
discussion.

On Mon, Aug 24, 2015 at 5:37 PM, Karen Coyle <[email protected]> wrote:
>
> On 8/24/15 1:57 PM, Tom Morris wrote:
>
>> What would it take to convince IA/OpenLibrary that a public bug tracker
>> is a critical part of the open source development process?
>>
>
> Tom, IMO the issue isn't about bug tracking but staffing and costs. There
> is virtually no staff time dedicated to OL, so the question of being open
> source and tracking bugs is somewhat moot. Even an open source project
> needs to make use of some human time. With the project essentially
> de-funded, having a "crowd sourced" bug list means that the bug list will
> grow, but few bugs will be addressed. It took something like two years to
> get OL re-indexed, and that seems like a no-brainer.
>
> As for it being open source, I'm not sure that is indeed the current
> thinking. Even if it were, it appears to me that the project lacks the
> tools needed to make code contributions possible (e.g. the test
> capabilities that I have mentioned before). Perhaps step one is to clarify
> whether the project is indeed intended to be open source (not just "source
> code lives in github"). That could help us manage expectations.


If the intention is for OpenLibrary to abandon it's 7+ year open source
legacy, that would be certainly be a much bigger issue than whether or not
there's a public bug tracker.  Having read writings by Aaron, George, and
other early OL leaders, it was certainly my impression that the choice of
being open source was intentional, not an accident, but I could be mistaken.

Cultivating a vibrant set of open source contributors could actually
*reduce* funding requirements, improve responsiveness, as well as overall
community engagement.  Ben, Charles, and others have already contributed a
number of bug fixes to OpenLibrary.  Willingness to promptly review, merge,
and deploy bug fixes would encourage others to contribute and provide a low
cost way to leverage scarce development resources.  A recent email
highlighted the 4 years worth (!) of community provide bug fixes which are
being ignored at https://github.com/openlibrary/bookreader/pulls. The first
of the epub issues, which have been around for years and years, finally got
addressed because two community members isolated and fixed the problem.

Public issue trackers aren't just for the technical community.  They also
help give the general OpenLibrary community which provides crowdsourced
metadata insight into the issues which hamper their work and the progress
which is being made to fix them.

Community engagement drives not only in-kind contributions of programming &
data-entry hours, but also real dollars for things like the recent year end
funding drive.  Engaged communities contribute more in all dimensions.
Disconnected and disenfranchised communities don't contribute in any way.

Grant writing organization also have a reasonable expectation that the
grantees will conduct themselves in a way aligns with their publicly stated
values and the values that they claim on their grant applications.  If
things like transparency, communication, public dissemination of open
knowledge, etc become perceived as buzz words which are paid lip service
rather than core values, it could easily reduce funding levels from
granting organizations.

On Fri, Sep 18, 2015 at 2:42 AM, Charles Horn <[email protected]>
 wrote:

>
> Here are my thoughts on the Jira and Github situation, I don't think
> having both is necessarily a sign of a problem in an open source project.
>

I agree.  The main problems are that 1) the change was made years ago with
no notice and 2) there's no way to access the canonical issue tracker
(Jira).


> Jira is a tool for tracking employee time spent on projects, development
> velocities and other internal and cross project metrics, not just issue
> tracking. It makes sense to me that the IA would have an internal system to
> track the work they are prioritising for devs who are likely working on
> many things, and not have this open to the public.
>

Jira can be used in lots of way.  If it's used to manage scrums as well as
track issues, access to the scrum boards can almost certainly be
restricted, if that's deemed important.  Jira also offers all sorts of
options for multiple projects, different security levels, etc.


> I have worked in places where we used Jira extensively and hosted open
> source projects on Github too. There is an extra effort to keep the two in
> sync if there is demand for it, but I think Github is always an appropriate
> place for community raised issues to be tracked, if the project is publicly
> hosted. It is up to IA to manage how that relates back to their internal
> systems. Hopefully community contributions are considered valuable and the
> extra effort of potentially double tracking in separate systems will be
> worth it for IA.
>

There are many different ways that things could be configured and at least
a few of them are likely workable solutions which would provide both a
shared communication space as well as provide for any perceived privacy
needs (although, I must admit, I don't understand the need for secrecy in a
non-profit dedicated to knowledge dissemination).


> Having an idea of IA priorities in regards to Open Library would be
> helpful, as would a clear set of contribution guidelines, but as a
> community we can put forward our own ideas too.
>

Yes, that would be great.  As would the ability to provide input into those
priorities (assuming that meeting the needs of the community is a goal at
some level).

I'd love to hear where the powers that be stand on this stuff. Is there
some kind of edict from on high that prevents communication about plans,
priorities, intentions, anything?

Tom
_______________________________________________
Ol-tech mailing list
[email protected]
http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
Archives: http://www.mail-archive.com/[email protected]/
To unsubscribe from this mailing list, send email to 
[email protected]

Reply via email to