launchpad-dev team mailing list archive
-
launchpad-dev team
-
Mailing list archive
-
Message #05666
Re: next technical architect kanban task
On Tue, Nov 16, 2010 at 3:54 PM, Robert Collins
<robert.collins@xxxxxxxxxxxxx> wrote:
> So, when I started this role, I put performance and continuous
> deployment - release features when they are done - as my kanban cards.
> These things are where I've been putting all my ongoing-project focus
> (vs day to day stuff like helping with design and code discussions).
>
> It seems to me that our continuous deployment project is nearly
> complete, and its time to start thinking about what I should take on
> to replace it.
>
It seems likewise to me. What's still to go?
> I've a few candidates already, and I'm interested in what all of you
> think would help you most. Thanks go to Gary who I've already run this
> past.
>
> The candidates I have already are:
> - fixing a bunch of baseline issues
> - test suite parallelisation: parallelisation initially, then
> parallelisation in our CI/landing robot
> - data mapping layer: Build and deploy across the whole code base a
> dedicated persistence layer.
> - faster/easier stock page creation: Suggested by Gary, this would
> aim to massively reduce the overhead in the non-ajax form development,
> so that we can focus on ajax form development. For instance, perhaps
> we'd get rid of generated forms and use node.js to render pages server
> side when working with a javascriptless browser.
>
> For now, I'm going to stay mum on which option I think has the best
> cost/reward tradeoff for us at the moment, and I'm sure I've missed a
> brilliant idea that will obsolete all of these :)
>
My money is on test suite parallelization, followed shortly by data
mapping layer. I think the slow test suite is a cause of a lot of
problems, and that either of those will contribute significantly
toward accelerating it. Of the two, parallelization seems like it
could deliver value sooner. I also imagine that most implementation
approaches would have the happy benefit of making it easier to set up
a machine for Launchpad development.
On "stock page creation", I'm surprised that we even care about
non-ajax form development any more. Can't we just skip to focusing on
ajax form development? (Maybe I'm missing something here).
I notice that you haven't mentioned two other hot topics:
* search
* use message queues for something in Launchpad (diff generation please)
Thanks for sending this through,
jml
Follow ups
References