RE: [PyMOL] Quadro-based 3D Mac Feedback

[Warren D. <wa...@de...>]

Joe,

Thank you for that candid and constructive criticism. =20

As a service-oriented company, our job is to advocate for and act in our
customer's interests.  When we encourage the adoption of one particular
solution over another, it is because we truly believe that our customers
will benefit from following our advice.

Through extensive travel, contacts, and personal experiences over past
few years, I have come to understand just how poorly Linux is doing at
meeting the needs of the scientific visualization community.  Despite my
unabashed pro-open-source stance, it would be unethical for me to
continue advocating the adoption of Linux as the general solution for
UNIX OpenGL visualization when in practice it is only effective for
those having strong system administration skills. =20

This is primarily due to incessant complications with proprietary OpenGL
drivers.  In my considered opinion, resolution of these issues is beyond
the reach of the open-source community, as no Linux vendor has
sufficient clout or motivation to effect the necessary and permanent
changes (i.e. shipping validated *proprietary* graphics drivers with all
common distributions of Linux).  In this case, the open-source community
is (correctly?) putting free software ideology
(free-software/open-source drivers!) ahead of users' needs to simply get
work done now.  In time, this approach may yield the desired end-result,
but scientists, doctors, and engineers need a working solution today.

Caught in the middle of all this, our obligation as a vendor is to
identify, advance, and promote solution which actually do work for our
customers today or very soon and not at some unspecified point in the
future.  We need an alternative answer to OpenGL under Linux, and we
obviously believe that answer lies with Mac OS X.

Up until now, another answer was Microsoft Windows.  However, Microsoft
has now made it crystal clear that Windows Vista will not support
current versions of OpenGL, and furthermore, that running older OpenGL
apps will compromise the quality of the Windows user interface
experience.  They are attempting to compel developers like us to adopt
their proprietary Direct 3D API, and trying to deny choice to users like
yourself.  This is simply outrageous and must be resisted.
<http://www.opengl.org/discussion_boards/cgi_directory/ultimatebb.cgi?ub
b=3Dget_topic;f=3D12;t=3D000001>

Apple, in contrast, is fully committed to OpenGL -- an open standard
that runs on any platform.  Furthermore, by switching to Intel
processors, Apple customers will now have even greater choice to run
one, two, or even three major operating systems on Apple hardware.
Simply put, Apple "gets it" when it comes to consumer choice, and over
the past several years they have made it a practice of synergizing with
Free Software and Open Source efforts (GCC, Darwin, KHTML, PyMOL, etc).
Steve Jobs understands that open-source is transforming the software
industry, but he also apparently grasps the new role that proprietary
solution vendors must adopt in delivering extra value on top of what is
already both ubiquitous and free.  That, in a nutshell, is the future of
the software industry.  Proprietary and open-source solutions will
co-exist and complement one another to the benefit of consumers and
developers alike.

> this blatant=20
> shilling contrasts jarringly with the open source context in=20
> which PyMOL is developed, distributed, and supported: PyMOL=20
> is attractive in part because it promises long-term, flexible=20
> access to the tools to do our work, without users falling=20
> into a vassal role with respect to one industry-dominant company.

I apologize for excess enthusiasm -- it is perhaps an overreaction to
biased and premature dismissal of Mac OS X as a credible solution by
some members of the pharma/biotech IT & Informatics communities. =20

> Launching such a hyperbolic cheerleading campaign may belie=20
> the stability of the support one might expect from Apple for=20
> this technology, if one feels such campaigns are necessary to=20
> keep that support alive.

But that is exactly my point:  If potential customers such as yourself
do not show strong interest in these systems through purchases or
communicate to Apple (directly or through us) why they may have missed
the mark, then Apple would be correct and justified in concluding that
we are uninterested in Mac OS X-based solutions for scientific
visualization.=20

In other words, if the nVidia Quadro FX 4500 is too high-end, but you
would definitely buy a $500 stereo-3D option, then you must say so.  If
you cannot see buying G5's now because of the pending Intel transition,
but would almost certainly buy an Intel-based equivalent to the Quad,
then you must make your voice is heard.  Likewise, if Mac OS X is still
missing critical software packages, then please name them.  Do not
remain silent yet unsatisfied.

> It seems premature, at best, to extrapolate from Apple's=20
> welcome, but very few and very recent, product releases to=20
> describe their support as "timely", "robust",=20
> "well-supported", "trouble-free" and=20
> "continuously-integrated".  Only with time can one make those=20
> claims fairly.  It seems they and their developer and user=20

We now have over three years of experience working directly with Apple
on PyMOL and associated systems.  They have been extremely responsive to
our concerns regarding UNIX source and X11 compatibility, Python,
Tcl/Tk, OpenGL bugs & performance, G5 and multiprocessor optimization,
and most recently Stereo 3D.  For a company with such diverse interests
as Apple to demonstrate this kind of commitment to the needs of an
supposedly niche community like scientific visualization is truly
remarkable. =20

Of course, I cannot say what they will do in the future.  All I can say
is that the people currently calling the shots with respect to
scientific computing are calling them appropriately.  That experience is
the justified basis for my strong endorsements -- Apple has earned them
well.

That being said, the core purpose of my email was to elicit feedback
from other who have not had such close and extended interactions with
Apple, and for us to understand the experience and perceptions of our
user community, especially with regard to these new visualization
workstations.  Some of the feedback already received has been quite
sobering.

So let me make clear:  We are not saying, "Go Mac!" blindly and without
reservations.  What we are saying is, "Consider Mac..." by carefully
evaluating their integrated hardware/driver/OS solutions and by sharing
with us your candid reactions to them.  If Macs are (still) unsuitable,
then please describe how they could be improved further.  However, if
they are now well suited to your needs, then please purchase them
aggressively.

I hope our position is now clarified, and offer my apologies for any
confusion, offense, or excess zeal.

Cheers,
Warren

--
Warren L. DeLano, Ph.D.                    =20
Principal Scientist

. DeLano Scientific LLC =20
. 400 Oyster Point Blvd., Suite 213          =20
. South San Francisco, CA 94080 USA  =20
. Biz:(650)-872-0942  Tech:(650)-872-0834    =20
. Fax:(650)-872-0273  Cell:(650)-346-1154
. mailto:wa...@de...     =20
=20

> -----Original Message-----
> From: pym...@li...=20
> [mailto:pym...@li...] On Behalf Of=20
> D. Joe Anderson
> Sent: Monday, January 16, 2006 9:34 AM
> To: pym...@li...
> Subject: Re: [PyMOL] Quadro-based 3D Mac Feedback
>=20
> On Sat, Jan 14, 2006 at 04:51:21PM -0800, Warren DeLano wrote:
> > The Quadro FX 4500-based PowerMac G5 is Steve Jobs' timely=20
> and direct=20
> > answer [...] The UNIX-based Mac has become the robust, low-cost,=20
> > well-supported, and high-performance replacement for SGI=20
> workstations=20
> > that Linux never delivered despite nearly a decade of=20
> opportunity and=20
> > effort (including our own).  Why does Linux fail with visualization?
> [...]
> > History shows that it takes an integrated hardware &=20
> operating systems=20
> > vendor like Apple (or formerly SGI, or Sun) to deliver and maintain=20
> > trouble-free OpenGL under UNIX.  Apple is the world's leading UNIX=20
> > vendor, and unlike with Linux, Mac hardware and operating=20
> systems are=20
> > continuously integrated with cutting-edge OpenGL graphics cards.
>=20
> > Has Apple in fact given us what you asked for?  If so, then=20
> have you=20
> > followed through with purchases?  If not, then what are you=20
> waiting for?
> > Is something crucial still missing?  Please share your=20
> thoughts.  We=20
> > guarantee that Apple and others will hear them.
>=20
> Despite the OpenGL workstation niche having been a yawning=20
> chasm for several years now, it took the concerted begging to=20
> get Apple to respond to that niche.  While congratulations in=20
> getting Apple to respond are still in order, this blatant=20
> shilling contrasts jarringly with the open source context in=20
> which PyMOL is developed, distributed, and supported: PyMOL=20
> is attractive in part because it promises long-term, flexible=20
> access to the tools to do our work, without users falling=20
> into a vassal role with respect to one industry-dominant company.
>=20
> Launching such a hyperbolic cheerleading campaign may belie=20
> the stability of the support one might expect from Apple for=20
> this technology, if one feels such campaigns are necessary to=20
> keep that support alive.
>=20
> It seems premature, at best, to extrapolate from Apple's=20
> welcome, but very few and very recent, product releases to=20
> describe their support as "timely", "robust",=20
> "well-supported", "trouble-free" and=20
> "continuously-integrated".  Only with time can one make those=20
> claims fairly.  It seems they and their developer and user=20
> communities will have their hands full with the upcoming=20
> hardware architecture migrations, during which support for=20
> many different niches might lag behind or fall by the=20
> wayside.  While one may hope they can pull it and keep it all=20
> together, let's not get ahead of ourselves: The list of=20
> Apple's predecessors in this niche, rather than being=20
> complimentary to Apple should serve as a sobering reminder of=20
> the difficulties Apple faces, and as a warning against=20
> hitching one's scientific productivity yet again to just one star.
>=20
> As for specifics of our own purchases, our purchase cycle had=20
> us replacing hardware early last summer, before the indicated=20
> hardware was available.  So, it doesn't seem to be a question=20
> of "what are you waiting for" so much as asking Apple "what=20
> took you so long?".  For this round, their support came too=20
> late, even though they've been trading for years on their=20
> claims of support for creative endeavors (cf publicity of the=20
> Genentech orders of the original G4 iMac).  Our next crack at=20
> it will probably come well after the migration to a=20
> Intel-based architecture is complete.
>=20
> As it is, we had to wait a year more than I'd have liked for=20
> funding, which put us in the position of having to navigate=20
> the replacement of hardware alongside a major release of the=20
> operating system (MacOS 10.4).  That brought with it=20
> additional support problems, notably a lag (or complete drop=20
> in) support for MacOS from other third-party developers,=20
> which issues we're still working through. =20
>=20
> Whatever the problems behind OpenGL support for Linux,=20
> entreaties to abandon it will not help convince Linux=20
> integrators or graphics vendors to support this niche.  In=20
> terms of choice and pricing, this will benefit neither Linux=20
> users nor Mac users.  Rather, what it does is re-animate the=20
> long dormant internecine Unix Wars of the days of old.  We've=20
> been through that, and we know how it turns out. =20
>=20
> For those who missed it the first time, it doesn't benefit=20
> Apple, nor users of any other Unix-like operating system.
>=20
> --
> D. Joe
>=20