[Linux4christians] OT: Soon, majority of users will no longer use IE

[AKAImBatman]

Rob, thanks for summarizing what I've been trying to explain! Linux really
has been doing well on the server side, but there's still quite a high
barrier to entry on the desktop.
On the issue of flat screen monitors, I'd say that the comparison is valid
as long as one understands what the tipping points are. Flat screen monitors
were superior to CRTs in nearly every way imaginable. The issues that held
them back were primarily price and service life. Once those two issues were
resolved, flat screens spread like wildfire.

In the case of Linux, the tipping point has nothing to do with price. Linux
is usually sold cheaper than competing options. Software availability is
definitely a tipping point that's holding Linux back. But if that was the
only issue, you'd think there would be more early adopters who use the
desktop despite the lack of software. Thus we get to the other tipping
point: Ease of use.

Linux desktops are very focused on technical competence. They can often do
cool things like auto-mount SMB/FTP/SSH shares without blinking an eye. They
can preview files on the fly, update the file size as it changes, provide
sophisticated and complex support for regular expressions, so on and so
forth. But if you ask them to do something user-oriented, boring, and simple
like create a main menu shortcut, they take a cue from Paper Mario and
comically fall flat on their face.

Which is a very curious issue. GNOME and KDE pour their time into some of
the oddest features. They spend time on "spatial" window placement as if
that will make the desktop 1000x more useful. They work together to create
cross-desktop shortcuts that only a small minority of users are going to
use. (Seriously, how often does the average user switch between GNOME and
KDE? Usually they commit to one and run with it.) In fact, the latter
"feature" was when the ability to graphically manage the main menu
disappeared from KDE and GNOME. (I can't remember if they eventually added
it back to the GNOME control panel or not, but I do remember that you had to
restart the desktop to get the icons to show up correctly. Meh.)

I'm not even going to go into the complexities of trying to install software
that isn't in the package management system. Something which is not only
easy to do on Windows and OS X, but is regularly performed by the average
user every day! (And for the love of all things holy, why are ELF binaries
still fixed to absolute file system locations, thus making a simple
untarrable/unzippable distribution difficult if not impossible? Gaaah!)

Thus ease of use is a huge tipping point for Linux, even 10 years after KDE
first appeared. If anything, the situation has gotten worse rather than
better. Back when I started with Linux, the system was an open book. Messy,
but open to whatever you wanted to do. And the GUI for the Red Hat Package
Manager did most everything it does today. (Though many of the programs
either wouldn't run or had no obvious method of execution after being
installed.) But we've managed to create so much pomp and circumstance around
Linux Desktops these days that it's actually harder to configure a desktop
that meets the needs of a user than it was back then. A very odd set of
circumstances.

(Of course, back then the underlying OS was so difficult to configure that
one was far better off running their favorite programs under FreeBSD than
Linux. That's one thing that has improved greatly.)

2009/2/5 Robinson Mitchell <robinson.mitchell at gmail.com>

> It really hasn't been my experience that Linux has been rejected by
> companies because of the free (as in lunch) factor.  Some years ago larger
> companies were reticent to deploy Linux because there wasn't a good support
> structure behind it that had any accountability.  That has changed.  Redhat
> built a major enterprise on the support and services model with Open Source
> free (as in speech) software.  Novell followed suit, becoming the second
> largest commercial Linux provider, using a similar business model.
> Again it wasn't the "free" thing; it was that businesses want accountable
> support options.
>
> I have worked at two large companies who had significant Linux
> deployments.  One of my prior employers, AutoZone, switched over 3,000
> servers from a proprietary OS on their store systems (SCO) to Linux
> (Redhat).  If you have kept up with Groklaw, you'll know that there were
> some repercussions of that choice that have not been resolved yet, but with
> SCO all but dead, I think that AZ is hopeful for a favorable outcome.
>
> A subsequent employer entered into an enterprise agreement with Novell and
> has deployed very significant server infrastructure on SUSE, replacing older
> proprietary Unix OS platforms - AIX, Tru64, and HP-UX.  Linux has been
> embraced in the server room and there is great rejoicing from the data
> center to the boardroom because it's providing cutting edge technology and
> saving big bucks.
>
> Ubuntu is a bit of a newcomer, but with its Debian heritage and the
> commercial support organization of Canonical behind it it has garnered a
> number of large server certifications, including becoming a certified
> platform for DB2 and Lotus Notes, both of which have pretty big footprints
> in the corporate space.
>
> In fact if you look at the data center space, Linux has made HUGE gains and
> outperformed the growth of Windows as a server platform for several years
> (read my words carefully - I said that Linux outperformed growth, meaning
> that the growth rate of Linux has been larger than that of Windows as a
> percentage of total deployments.  This does not mean that more Linux was
> deployed than Windows. This just reflects the fact that the footprint of
> Linux in the Data Center grew as a percentage of total platforms, and
> Windows lost market share in the server space.)
>
> On the desktop Linux deployment has not cut significantly into Windows
> market share, not as much as Mac has.  This is mostly owing to the
> overwhelming predominance of Windows as a business desktop platform and
> compatibility concerns.  The big issue is that the vast majority of line of
> business applications are Windows-based.  The lack of market share becomes a
> Catch-22 for software vendors:  there's not sufficient Linux market share
> for them to have incentive to port apps to Linux, and the market share
> doesn't grow because there aren't a sufficient number of business apps
> ported over.  This conundrum is, I think, the primary reason businesses have
> been slow to adopt Linux.
>
> Most of us as Linux users are accustomed already to using the
> business-ready universal applications like Open Office, but think of they
> myriad of niche products that exist for specific types of businesses, for
> which no Linux app really exists.  The lack of specialized line of business
> applications is the achilles heel for Linux, in my opinion - it's not the
> OS.  It's the app space that's the real problem, at least it seems that way
> to me.
>
> That being said, the "Free" thing does affect the issue, in that line of
> business applications are not forthcoming because these sorts of
> applications are generally produced by for-profit companies.  Yochai Benkler
> [The Wealth of Networks]  has described the emergence of different models of
> production, foremost of which is the community that produced Linux itself -
> a marvelous achievement.  It is going to take time for niche
> line-of-business applications to emerge in the same fashion, not necessarily
> because of the "free" thing, but because only a few companies are savvy
> enough to figure out a business model that allows them to monetize
> production of free software.  A number of business models are emerging, but
> not all are successful.
>
> I do think the economic downturn will spur a recommitment to development of
> niche applications in the Linux space, however.  It remains an open question
> how long it will take to reach a critical mass - but Linux desktop apps are
> not the only thing that is developing in this way.  Think of an example that
> has had to go through the same process:  flat screen monitors.
>
> I won't go into detail here, but if you will refer to Malcolm Gladwell's
> book The Tipping Point, he describes the phenomenon well, and he uses flat
> screen monitors as an example.  It's a good one.  It took time to go from
> first prototype, to first market penetration, to the point that flat screens
> displaced CRTs, but it is a predictable pattern.  Linux Desktop apps have to
> emerge in the same way.  The process will be slower because the number and
> variety of line-of-business apps is so much greater than the number and
> variety of types of flat screens.
>
> Oops, this post got quite a bit longer than I meant, but it's something I
> think and read about a lot. It is not something that lends itself to
> superficial analysis and assumption.  There is no sound bite answer.  This
> is one we have to wrestle with.
>
> Kind regards,
>
> Rob
>
> _______________________________________________
> Linux4christians mailing list
> Linux4christians at thelinuxlink.net
> http://www.thelinuxlink.net/mailman/listinfo/linux4christians
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.thelinuxlink.net/pipermail/linux4christians/attachments/20090205/5312b4dc/attachment-0001.html