launchpad-dev team mailing list archive

[Gary Poster <gary.poster@xxxxxxxxxxxxx>]

On Jul 26, 2010, at 6:21 AM, Jonathan Lange wrote:

> On Mon, Jul 26, 2010 at 11:16 AM, Robert Collins
> <robert.collins@xxxxxxxxxxxxx> wrote:
> ...
>> I guess, to put this in strategist territory :P - you can have any two
>> of : fast, correct, cheap(dev); right now we have correct, cheap(dev),
>> but the message I'm hearing from our users and stakeholders is 'fast,
>> cheap(dev)' : you seem to be saying 'fast, correct' :- we're not
>> positioned to do that yet.
> 
> I want users to be able to choose from:
>  fast, cheap(dev)
>  correct, cheap(dev)
> 
> And I'm happy for the first

From the following sentence, was this a typo--did you mean that you were happy for the *second* ("correct, cheap") to be the default?

> to be the default. Users who want fast &
> correct can figure it out for themselves.
> 
> Of the options you list above, the first two sound as if they are
> sacrificing cost of development for correctness and speed.

I think that Leonard's current approach of making len be something that we can code to be requested as a separate request, if we know it will be expensive, is the right compromise for the webservice.  I think this is in line with what Jono is saying.

Moreover, even if Robert's idea were decided as a way forward, I'd rather not schedule it for several months.  We just finished a performance focus on the webservice (https://dev.launchpad.net/Foundations/Webservice/Performance) and, FWIW, the current webservice focus plan is the following.

(0) We were told that helping Soyuz with the len issue was critical, so we pushed this to the front.
(1) Once that is done, we'll return to the desktop integration work.
(2) Once that is done, we'll move to the long-planned focussed work on the webservice's usability.

Certainly performance is an aspect of usability, but there's a lot of other stuff to do there too.  I'd rather return to performance after following the plan we have already.

Gary