Thanks for your work on the web site.
Your Github respository is not a fork of www-sugarlabs repository; why
is that? Was there some advantage in making a new repository?
Are you still planning to target the laptops that run Sugar? I ask
because your design has gone backwards two orders of magnitude in
terms of performance and load time. Your design on 14th March was 6MB
over 35 requests from one domain, but your design now is 12MB over
153 requests across many domains.
Without a test harness for evaluating the impact of each change, and
continuous integration, my guess is that you've allowed changes that
will make it even harder for you to target the laptops that run Sugar.
So I do not think it will be possible for you to target the laptops
that run Sugar, and it will be best to have the Browse activity in
Sugar not link directly to the main page. If you agree, please make a
subpage for use by Browse and make a pull request against Browse, file
data/index.html which we can hold until the site goes live.
On the other hand, if you would like to target the laptops that are
running Sugar now, you might do these things;
- install a Fedora 18 virtual machine and use the Epiphany web
browser, which is based on WebKit,
- install an Ubuntu 16.04 virtual machine and use the Epiphany web
browser, which is based on WebKit2,
- use the virtual machine configuration to restrict the resources
available; such as the VirtualBox Settings System -> Processor ->
Execution Cap value,
- use a virtual machine API to record CPU usage during tests,
- set up continuous integration test harness that will wait for a
git commit, load the page in fresh browser, and measure CPU time
Hope that helps.
I have posted my GSOC project report for the 2nd week on my blog. My blog can
be viewed at https://psramarajugsocblog.wordpress.com. Do check it out and
provide your feedback.
Thanks and Regards,
Sugar-devel mailing list