[MeeGo-dev] browser engine and other big architectural decisions

[Andrew Flegg]

2010/3/24 George Matveev <geoartmat at yahoo.com>
>
[snip interesting insights]
>
> The end result - very poor browser performance on all Tablets
> (Loading objects, Loading objects, with complete freeze of UI),
> frustrated customers and ultimately the failure of the platform
> (since for most users browser is most popular application).

However, every review of the N900 says it has the best mobile browser
by *far* and that it's the one redeeming feature for all the other
shortcomings. Personally, I'm often frustrated by Maemo browsing but I
think this is for three reasons:

  1) Small screen and websites not designed for it
  2) Poor bandwidth performance due to device or available 3G network
  3) Poor browser engine performance

#3 is apparently tackled somewhat in Maemo 5's PR 1.2 which includes
an update to the Gecko core and numerous performance improvements.

> To become a success a mobile stack needs to select best open source
> components and integrate them seamlessly and effectively on best
> platform (i.e. Debian) . LiMo is already doing this.

I think pointing to LiMo as an example of a successful, open, Linux
stack is somewhat ironic - if LiMo was so successful there'd've been
no need for Maemo, Moblin or even, now, MeeGo.

However, I don't think anyone would disagree with your assertion of
"select[ing the] best open source components and integrat[ing] them
seamlessly and effectively on best platform". It's the detail as to
what "best" is where people disagree (rpm/deb, yum/apt, Gecko/WebKit)
;-)

Cheers,

Andrew

--
Andrew Flegg -- mailto:andrew at bleb.org  |  http://www.bleb.org/