My understanding is that the problem with Sun offering a SunRay software
client is support: As soon as folks try to download and use that client,
not only will they encounter issues with the SunRay client itself, but
also with running that client in the user's environment, e.g., Windows,
etc.
So by offering that client, Sun would have to start supporting an app on
Windows (and Linux, etc), which, as we all probably know, is no fun, and
is quite
expensive. In fact, for users, this would also circumvent the whole
operating cost
reduction issue that's one of the SunRay's biggest selling points.
I don't think it's realistic to expect a commercial entity to release a
product without also offering some form of support for it. But that,
I think, is where open-source really could play a role. If such a client
were developed in an open-source fashion, few users of that software
would expect Sun to support it. Instead, the open-source community
around that software would support it. Actually, that would play well
with the likelihood that such a client software would initially be used
by sophisticated early adopters (like folks on this list), who would
actually be able to contribute back to the community.
One possible business model for Sun might be related to their display
grid initiative, which is a service of remotely-hosted SunRay sessions.
I've been trying out a prototype of that service for a while, and it's
very impressive so far. If Sun can sign up a large ISP that would market
that service even to residential customers, that could generate a lot of
interest in the SunRay.
At that point, having a software-based, open-source client would help
people get a feel for the service. But the broader issue then is to
open-source much of the SunRay software itself so that others can write
all kinds of interesting services around that software (the software
client being only one possibility). Much of that software would be
useful only in connection with having a rock-solid, remotely available
"grid" back-end that serves up the SunRay sessions (and other grid
services), and Sun's grid initiative is really aimed to provide such an
infrastructure. If the display grid idea takes off (and I think this is
only a matter of time), SunRay DTU prices will probably fall, just as
router, wireless router, etc. prices have fallen with a growing market.
So selling DTUs won't be much of a business opportunity (unless maybe
selling millions of them at a time to ISPs), but selling the HA grid
services will be, I think.
Ken Mandelberg wrote:
First, to the narrow issue of the original question, a softray client
for Sun Solaris/Sparc workstations. It seems to me that whatever the Sun
business model is, they can only make more money selling the workstation
than the Sunray. The fact that we can't move a Sunray session to a Sun
workstation is driving us to buy cheap Sunray's instead of expensive
(but more flexible) Sun workstations. We at Emory University have been
pressing this issue for year with no luck.
Second, to the more general question of a softray client for non-Sun
hardware. Sun has apparently decided that it has a viable business model
giving away Solaris for other vendors hardware, and selling the support.
In fact with OpenSolaris source code also free, its easier than before
to avoid even buying the support.
If the business model for Solaris is suppoed to work, why wouldn't it
also work for Sunray?
There are constant statements from Sun executives about open sourcing
more and more of Sun's software portfolio. Is Sunray the exception?
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
