Dear Faidon,
Sorry for not responding sooner to your thoughtful recommendations about next year’s plan.
We have been swamped this week, to address a flood of community responses to Media Viewer on Enwiki, and are just now catching up with previously scheduled tasks. :)
To answer your question below, the ‘Critical bugs’ category intends to cover two main activities: critical bugs and technical debt. We are allocating about a third of our time to this category next year, evenly split between these two activities.
We have purposely kept that category open-ended, so we can determine every few weeks which specific tasks to take on: for example, when image scaler issues started causing problems for the operations team, we switched our focus to work on solutions to those issues.
So if the thumbnails issue you describe below turns out to be the most critical issue to tackle next, we would turn our attention to that problem, using up that 30% of team capacity to attempt solving it. Over time, we hope to address some of the most important issues with our aging media infrastructure.
Gilles will be spearheading that technical initiative, in collaboration with Mark and Gergo. He is on vacation this week, but can respond in more detail then.
In the meantime, I hope this first answer can give you a sense of where we’re headed.
We look forward to working with you to make progress on these fronts in the coming year.
Regards as ever,
Fabrice
Fabrice & team,
A bit past the goals deadline, but I hope I can still be useful :)
On Mon, Jun 02, 2014 at 04:59:05PM -0700, Fabrice Florin wrote:
3. Activities
This year, we propose to invest our time on these main activities:
• Structured Data: improve the way we store metadata — an important goal, because most other features depend on it.
• Critical Bugs: fix urgent bugs in our infrastructure -- and become more familiar with our entire code base, including:
- Image scalers, Core media handling, TimedMediaHandler, Media backend storage, Media Viewer, etc.
• Features: develop or improve user-facing projects, including:
- Upload Wizard -- our main user-facing project this year (includes Commons Upgrade and Modal Tool for other sites)
- Other Features -- Media Viewer 0.3, File Notifications, File Page, Kaltura Player Upgrade, Campaign Tools
I notice that you've split the roadmap into "critical bugs" and
"features". What exactly do you mean by "critical" here? Is there time
allocated for non-critical but also non-feature work, e.g.
infrastructure/architecture design?
Our current media infrastructure is getting old, has been traditionally
under-resourced and hence has accumulated technical debt over the years.
As an example, there are currently four RFCs related to thumbnails and
I'm sure opinions on their characterization as "critical" would vary.
Additionally, as you know, we've had some outage lately that can and
have been dealt with by deploying hotfixes, but unless we invest time on
a properly scalable & secure architecture these issues will keep
cropping up. Our thumbnailing architecture could surely use some love :)
My impression is that the team has been doing a bit of both, as time
permits; is this the plan going forward as well? I've discussed various
ideas with the team (mostly Gilles) and I know there is both the will
and the necessary combination of skill and experience to do those larger
changes, so I'm wondering if there's going to be time allocated for
this, this upcoming FY.
Thanks,
Faidon
_______________________________
Fabrice Florin
Product Manager
Wikimedia Foundation