On Wednesday, 8 April 2015 07:28:05 CEST, Karan Luthra wrote:
I will be talking about Trojitá with the aim of introducing what the application does, how it does it, and where it needs further development.
Hi Karan, this sounds awesome. Thanks for doing this :).
At this point, any suggestions about the content of the talk are very welcome. Any past experiences or blogs/video footage of talks that went well and those that din't might be helpful to me.
I've only given two talks about Trojita. The first one was more like a demo of the N950 version, there were no slides and there's no video. The second one was in Czech at my MSc. thesis defense. There's no video and the (Czech) slides are in Git, docs/talks/2012-09-03-msc-defense.* . That talk focused on the IMAP theory rather than on Trojita-as-an-application for obvious reasons.
I don't plan this 30 minute talk to have any hands-on dimension to it, (might utilize some other time slot, gathering interested people) so *any* suggestions on making the talk interesting and exciting for college undergrads are welcome. If I had to define the goal in one line: after the talk, people should be able to look at the bugzilla and know what areas to pick up a bug from. And also know that there are a lot of areas in Trojitá that could take contributions.
That sounds good. There's a couple of messages which we repeat a lot on the web, such as "Trojita is fast". Many people are used to ignoring these "advertisement bits", so when they try it out themselves and find out that yeah, Trojita actually *is* fast, they are rather impressed.
So I would probably start by doing just that. Use Trojita built in the release mode, not debug mode, point it at a local Dovecot instance on your laptop with a mailbox that has 50k or 100k messages, open that mailbox. Be sure that Dovecot has proper indexes, we want to show how fast Trojita is, not how long it takes for Dovecot to reindex the mailbox, and Dovecot looks at what clients are using to see what to include in its indexes. Open that mailbox again, now that Trojita has the data in its cache, to show that it only gets faster, not slower.
That should get some people interested, I hope. You might want to compare this to other e-mail clients, or even ask how long it takes them to resync their 50k mailbox with their favorite clients :). A problem with this strategy is that webmail users won't really care because this is effectively hidden for them by e.g. GMail.
You might also mention that the application scales; it is used on mobile phones, on deksopt systems, and in an ERP/CRM business platform, too, using a single code base.
We have a nice CI system which informs contributors about build problems and test regressions on platforms which they might not have access to. This means that people shouldn't be afraid of sending patches, and that we should not be afraid of merging stuff because we do have seatbelts in place here.
We also have a nice and friendly development team of course :) and we generally stopped eating clueless first-time contributors for dinner a couple of years ago. We can promise newcomers that their patches will be reviewed and that the feedback which they get back will help them improve the code, not make them feel bad. If they're looking for feedback from a real-world active and not-so-small but also not-overwhelmingly-large project, Trojita is probably a good stop.
I'll be of course happy to give you some feedback on your slides if you wish to have a rehearsal or something.
Cheers, Jan -- Trojitá, a fast Qt IMAP e-mail client -- http://trojita.flaska.net/
