On Thu, May 15, 2014 at 8:35 PM, Eric Meyer <
xor...@gmail.com> wrote:
> dmccunney wrote:
>>
>> Why should they? They never need to deal with plain text.
>
> Need? Why wouldn't they want to? Much of the time text is just
> information, and there's never going to be printed output, so no, appearance
> is not a factor, and I don't want it in some clumsy proprietary format that
> fusses about that (and will be hard to translate a decade later).
Think email. While I'm one of those who prefers plain text for email,
many or perhaps most don't The dominant format for email these days
is HTML, which lets you do things like use color, fonts, embedded
images, and hyperlinks. Your email client interprets the HTML and
displays the formatted message on screen the same was a browser
displays a web page.
People really do care about stuff like that, even if it arguably is
cosmetic and not intrinsically necessary for communication. The nadir
is probably the packages of animated smileys you can use in email, but
I tend to not exchange mail with people who use such things. We
simply have nothing meaningful to say to each other.
HTML is not a proprietary format, and proprietary formats are slowly
going away. Current version of Microsoft Word, for example, store
documents in XML, and the underlying document is plain text. Word
parses the XML and obeys the formatting instructions specified in it,
so what you see is colors, fonts, text attributes, embedded image and
the like. The old proprietary format was simply a different and more
opaque way of doing the same thing: specifying what the document
should *look* like. (Of course, since it's Microsoft, their version
of XML diverges somewhat from the official standard, and they caused
some fuss a while back by trying to push their version *as* the
standard.)
>> The intentions behind Australis were good, but they don't seem to have
>> checked to see how existing FF users would feel about the new UI.
>
> Why mess with a new UI? Who is it for? The problem isn't that Firefox now
> looks like Chrome, though that does seem stupid, but that a whole new UI
> installed itself automatically like just another incremental update (unless
> you prevented it). I find it hard to believe that any users would want
> that.
Firefox went to a rapid release model a while back. Instead of
occasional big releases that added a lot of new features and fixes,
they went to a new version every three months, with fixes and less
changes. By default, you get new versions and any updates to addons
you use automatically. Google Chrome does the same thing. So does
Opera, though Opera now uses the same open source rendering engine
Chrome does.
Australis is actually the first major interface redesign in a long
time. The last really major change to Firefox was the transition from
v3.X to 4, because there were changes to the underlying architecture
that broke an assortment of existing extensions. The extensions
relied on the way FF used to do things, and stopped working when that
changed. FF still *looked* the same, but what it did under the hood
was different.
The Australis redesign has it roots in the increasing migration to
mobile devices. People are using browsers on notebooks, netbooks, and
tablets, and the scarce resource is screen real estate. The goal of
Australis was to simplify the interface and reduce the amount of space
taken by the browser chrome to leave more room for what you were
browsing.
Australis is intended to look and act pretty much the same on any
device you run FF on. I have it on an Android tablet, and it's quite
similar to what I see on my desktop.
It doesn't perturb me as much as it perturbs others because I know how
to make it look and act as I prefer.
> I don't want to turn this into a Mozilla thread (there's already a ton of
> rants about this online)... I was just using Australis as an obvious recent
> example of unnecessary, disruptive change. The waste of the developers'
> time is dwarfed by that of the users. I expect that from a company like
> Microsoft, but why should it happen now with open-source software too?
It's always been a problem with open source software, and a persistent
complaint about open source offerings is poor UI design.
I think the issue has a lot to do with the relationship between the
developers and the users. In commercial software, developers are paid
to work on the products, and the money in their paychecks comes from
sales of the products they work on. There is a strong incentive to
pay attention to customers and attempt to provide what they want so
that they *buy* the product. Developers don't code what they feel
like - they code what their manager tells them to code.
In open source, the relationship is very different. Most developers
working on open source products don't get paid for it. They do it as
a sideline activity, and make their living writing commercial
products. Generally speaking, if you get paid for writing open source
code, you work for a company like Google, Facebook, or IBM that uses a
lot of open source code and pays you to work on what they need it to
do. Those not getting paid for it are either scratching a personal
itch, working on code that they use, or are in it for community
status. Things like "I have commit access to the Linux kernel git
repository and Linus accepts my changes to the code" are major status
markers.
So you get what Jamie Zawinski called the "teenage attention deficit"
model of coding. Jamie has been around a long time. He was one of
the Netscape developers back when Netscape was an independent company,
and one of the original employees of Mozilla when the Mozilla project
was spun off from AOL. Jamie was fulminating about the changes in the
Gnome desktop environment that was widely used in Linux. Gnome 2 had
a list of unfixed bugs as long as your arm. But fixing bugs isn't
*fun*. Writing *new* code is fun. So instead of a properly fixed
version of Gnome 2, we got a redesigned Gnome 3, with a whole new set
of bugs unlikely to get fixed. The developers were not responsible to
the users, and had little incentive to care what the folks who will
actually run their code think. They were writing what they wanted to
write, and end user satisfaction wasn't a criteria in what they did.
Mozilla has an overall mission to improve the web and develop web
standards, but that isn't quite the same thing as making users happy.
The problem with the Australis UI redesign is that it was conceived in
a vacuum as a needed change to position Firefox and other Mozilla
products for the future, but there doesn't seem to have been any
effort to reach out to the existing user base, explain what they were
up to, and get feedback on what the actual users wanted. We don't pay
for Firefox, so we lack the leverage of voting with our wallets and
refusing to buy if we are unhappy with the direction things are going.
I don't know of a good solution to the underlying problem
> -- Eric Meyer.
______
Dennis
https://plus.google.com/u/0/105128793974319004519