[blfs-support] browsers (was Re: Latest news in GNOME world)

Simon Geard delgarde at ihug.co.nz
Mon Nov 19 23:56:35 PST 2012


On Mon, 2012-11-19 at 08:41 -0500, LM wrote:
> A lot of the major people involved in the WhatWG also seem to be the
> major developers involved in developing browsers.

And I think that's how it should be. Along with the content producers,
browser developers are the people with the biggest stake in what a
standard should look like - there's no point in producing a
theoretically-perfect specification that can't actually be implemented
in the real world.


> Am not at all thrilled with the release early, release often mindset.
> From what I've read, Mozilla is doing this to keep up with Chrome.

It's nothing to do with Chrome - it's more about keeping up with the
spec, and with pushing features out as quickly as possible. It's about
not being like Internet Explorer, always being several years late in
adopting new technologies.

I'd also note that a lot of this is being driven by mobile development -
by Apple and Google in iOS and Android respectively. Mobile devices have
a bunch of capabilities that traditional machines don't - cameras, GPS,
etc - and web developers want to take advantage of those things.


>  A major issue is that online tools such as Google Mail will typically
> only support 2 versions back in software.  However, big corporations
> and some web developers like IBM (we use their Cognos BI tools at work)
> can't afford to keep up with this development and release pace.

Yeah, that's a problem we had to deal with at work, when they switched
to the rapid release cycle. But in practice, what we've ended up doing
is simply ignoring it - our experience is that we can upgrade Firefox
whenever a new release comes out, with full confidence that nothing will
break. The move to out-of-process plugins in Firefox 3.6 was the last
time I remember anything causing problems for us...

And so we've found it's mostly a theoretical problem - as long as we
know what the oldest version our clients could be running is (and we
*do* know), we simply code without requiring features added since then,
knowing it's not going to break if the client has auto-update turned
on...


> There's also so much breaking away from standards or
> lack of implementing them in the available browsers that for a web
> design to look good on multiple browsers you have to special case a
> lot of things.

I don't see that. As you said earlier, the standards are being *written*
by the browser makers, and most of them (Safari, Firefox, Opera, Chrome)
are keen to implement the new features. True, support for some things
does vary between browsers, but that's a matter of timing rather than a
refusal to implement things.

>  Now, HTML 5 is the latest bandwagon and I run across many sites that
> say they're specifically designed not to work on IE.  You would think
> the end goal of a web design is to reach every potential reader,
> customer and/or client out there, not to limit it to a priviledged few.

And I *definitely* don't see that. There are sites that don't work on IE
because it lacks features web developers want to use, but not because
developers have deliberately excluded it. Or is that what you mean?


I'm trying not to beat up on Internet Explorer too much, but a lot of
compatibility problems come down to that browser being very slow to
support new features. It's their long release cycles - other browsers
can implement a new HTML5 feature and have it shipped in a few months,
but when Microsoft only ships a new version every couple of years, it's
impossible for them to deliver new features promptly...

Simon.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://lists.linuxfromscratch.org/pipermail/blfs-support/attachments/20121120/63c203bd/attachment.sig>


More information about the blfs-support mailing list