[geeks] Browser licensing?
Joshua Boyd
jdboyd at jdboyd.net
Wed Sep 3 17:39:59 CDT 2008
On Wed, Sep 03, 2008 at 06:11:18PM -0400, Alois Hammer wrote:
> But, anyhow, Google's modified WebKit (and the "V8" JIT engine?) are
> licensed under BSD. Except that they're not. Wikipedia claims BSD, but
> http://code.google.com/p/chromium/ says "Code License: Multiple
> Licenses." What's that mean? Is that the default for when there's no
> registered license yet? Does it mean LGPL/BSD? No one's said anything
> about any part of Chrom[ium] being licensed under the LGPL, which sounds
> like either a license violation, or else "we rewrote those parts
> ourselves so everything could be BSD."
>
> ...and there's no CODE at the site. So, unless Google's chosen to
> distribute the source via some other means than their mighty
> project-hosting site, no code's been released to anyone outside Google.
> If there's no code release, it hasn't been licensed. To anyone. As
> anything.
What is so hard about:
wget http://build.chromium.org/buildbot/archives/chromium.tgz
Be prepared to wait a long time though because it is 458 megs of
source and the server seems to be a bit slammed. They helpfully mention
that it expands to over a gig and that you'll need 10 gigs free to
actually build it.
Also, numerous pieces are available individually via SVN, such a V8 and
the Skia graphics engine.
FWIW worth, it is obviously multiple licenses for Chrome. V8 alone
has multiple licenses.
> Ignoring the impossible performance claims,[1] the 0.2xx version
> numbering (which, from reviews, I'm taking at face value), the
> catfighting on full-disclosure, and OSNews pointing out that this looks
> very suspiciously like IE8b2 but not as good-- this is not an auspicious
> start.
>
> ---
> [1] I can't wait until people figure out that instantiating an entire
> JSVM for anything other than, oh, Google Reader is actually a
> performance *loss*, especially compared to Gecko JS tracing or WebKit's
> SquirrelFish. (Or SpiderBaboon or DolphinElephant. Whateveritis. These
> names are getting tiresome.)
Why must instansing a new VM be slow? How is it impossible for someone
to figure out how to do it quickly?
More information about the geeks
mailing list