ConFoo 2012
27 Feb - 3 Mar 2012
at Hilton Montreal Bonaventure Hotel
Talk comments
Guilherme, given the scope of the talk, the fact that organizers suggested not showing code because of the size of the room, plus Hugo being there providing very detailed Symfony information I decided not to spend too much time on Symfony2 or our implementation. Anything I covered would require an understanding of how Symfony works so I would lose a lot of the audience.
We have buy in on from high for giving back our improvements. Often they are too unique to be of interest to the general public. For example our Redis service is very tiny and only supports the features we needed, it isn't robust enough to be of interest to anyone else yet.
Developers from Manwin have posted patches to JQuery and other open-source projects though. Not all devs use their Manwin email addresses when doing so but it happens and is supported.
Personally, i didn't like the approach used by the presenters. For me, they look like 3 guys that needs to fill 2 days on PHP concepts. The worst part was when they were arguing on something like how to put the braces in the code.
I wouldn't go to this training next year.
Link to the slides:
http://www.slideshare.net/starbuck3000/confoo-2012-web-security-keynote-11979308
Can you share your slides :) ?
Eric,
Excellent job; you solidified a few things for me with this talk that I had been contemplating. Thanks a lot for giving back by sharing your experience. The community will be better for it.
BTW, congrats on re-proving that high-traffic sites/applications are just fine with correctly architected PHP.
Certains éléments sont à la jonction de plusieurs critères mais je pense que les éléments temporaires appartiennent au critère d'exhaustivité car leur présence indique l'absence d'une partie de l'application finale.
Wil, thanks. My goal more than anything was to show people that there are software packages beyond the standard LAMP base that offer great benefits if you spend a bit of time on them. While it wasn't the best fit for our requirements, MongoDB is also excellent.
If our requirements change, there is no reason why MongoDB might not become part of our stack. I know other projects at Manwin are experimenting with it. I'll always use the best tool for the job and not try force a tool I know to do things it wasn't designed to do.
We are always experimenting with interesting new technologies that might help us. A lot don't scale or are too buggy. But we try them all. Hopefully we can continue to give back useful information about what we learn.
Given that Facebook is written in PHP, I doubt that PHP really needed any more proof IT'S scalability but hopefully I helped with some ideas about other techs that we have learned do scale.