Interview with Lewis Rosenthal
TITLE: Interview with Lewis Rosenthal
DATE: 2014-07-25 18:52:28
AUTHOR: Lewis Rosenthal
Q: Do people recall the word "IBM OS/2 Warp" in USA today?
Some older folks recall the word. Mostly, the younger group confuses "OS/2"
with "OS X," which is quite funny.
Q: What is the area of OS/2 usage in 2014 and 2015? What do you expect?
I think that Microsoft has opened the world to alternatives with their
fumbling of the Windows ball. I have more and more clients willing to look
at alternatives for what they do (being pushed off of Windows XP, they
realize that whatever they choose is going to be considerably different from
what they've become accustomed to using). I think that OS/2 - or some
derivative of it, whether eComStation or something else - becomes one of
those alternatives to consider.
Interestingly, as more people consider alternatives to Windows, Microsoft
Office becomes less and less a "must have" in business environments,
competing alongside LibreOffice and OpenOffice. Once businesses realize that
they no longer need to maintain their Microsoft investments, a world of
opportunity opens before them. We have a way to go before we are
automatically considered as a desktop replacement to XP or Windows 7 in such
environments, but we're getting there. An updated OpenOffice, and the
approaching end of Windows 7 mainstream support January 13, 2015 would make
an OS/2-based solution much more viable. (Note that extended support for
Windows 7 is scheduled to continue until 2020, however, that does not mean
that new or updated applications for Windows will necessarily support
Windows 7 throughout its extended lifetime.)
Home users are still likely to have some preference for other platforms,
simply due to our lack of comprehensive multimedia support. However, this
too may change in time, as people depend more and more on tablet devices for
Q: How many developers should work on the operating system core to get results?
I think a dozen developers would be a good start. These would be focused on
kernel enhancements, networking updates, and device driver development to
support newer technologies. Of course, that is not to day that 50 or 100
developers wouldn't be a good thing, but I'm being realistic about the
project and about the available resources we have in the community at large.
It's also not just about development. We have support issues to address.
People pay for support and updates, fixes, and enhancements which are short
of a full operating system upgrade. That coding also needs to be considered
part of the OS "core." Tied into that, of course, is the need for reliable,
available vendor support, with support upgrade opportunities (fee-based
support beyond bundled support). Unfortunately, like development, this too
requires human resources, not only to provide the support but to train the
support staff. I haven't addressed these human resource requirements in
answering your question, above, which only points out what I feel we need to
move the OS forward.
Q: How to develop OS/2 without financial resources? What method to use?
It is literally impossible to provide proper development energies for the
operating system without some form of professional labor. Whether this is
paid by venture capital, funding angels, or just mass contributions by the
current user base, good programmers need to be hired, not conscripted for
volunteer work. Surely, there is are roles for unpaid volunteers, but
quality code comes at a price. A hallmark of OS/2 since its inception has
been it's amazing stability. That stability comes at a cost.
So, to answer your question, it would not be possible to continue
development without financial resources.
Q: What system components can be (must be) replaced?
Disk drivers need to be updated to better support SSD partition alignment.
Network drivers (wired and wireless, though wireless drivers are extremely
important) need to be constantly updated. The video driver (whether SNAP or
Panorama - and SNAP is almost a dead horse, at this point - needs to be
updated and enhanced. USB 3 support will become more important in the near
future. Audio support also needs to be enhanced.
Now do any of the above constitute "replacement?" That's a matter of
opinion. They all constitute maintenance and enhancement, however. MultiMac
is (or will be) a "replacement" for GenMAC, which was never very
supportable, anyway (I wrote in May of 2007 about the need for a replacement
for GenMAC, and was roundly criticized for my opinion:
Q: What is the main problem of OS/2 Warp / eComStation today?
I think the greatest "problem" we as a user community face is the same thing
which has plagued us since the beginning of OS/2: lack of adequate hardware
support. Hardware manufacturers do not have enough incentive to code the
necessary drivers for our niche market, and many are against releasing
enough information about their devices (supposedly for fear of others
infringing upon their intellectual property) to allow for interested third
parties to develop good drivers (and this is something with which the Linux
community must contend, as well: look at some of the better video cards on
the market today, and manufacturers like nVidia, who until recently have
been slow to embrace open source development of drivers for their cards).
Q: Is now a good time to Virtualize OS/2 and/or eCS?
Virtualization is a mixed bag. For situations where there is a single OS/2
application which hangs on, and no supported hardware, virtualization
extends the life of the code, allowing the application to continue in place.
However, for those who use many OS/2 applications, virtualization can become
an unnecessary layer of overhead, slowing the overall system, and reducing
its usability to the point of forcing the user to switch to another - native
I have clients who run entire business operations on OS/2 and eComStation,
some with one or more custom applications, running across literally hundreds
of machines. The thought of virtualizing each one of those installations is
truly daunting, and supporting two operating systems - even for a single
application - simply because of all that that application does, seems like a
huge waste of resources.
I use eCS as my daily operating system. I use an OS/2 browser (SeaMonkey),
mail client (also SeaMonkey), application suite (OpenOffice), and other
management tools which allow me to access other systems from my Workplace
Shell desktop. I routinely use other platforms, but I am most at home with
eCS. I don't want to switch to another platform. For me, the killer app is
the WorkPlace Shell itself. Unfortunately, I am running headlong into an
inevitable conclusion: my aging ThinkPad which is going to need replacement.
The newest ThinkPads do not fully support eCS. Running eCS virtualized in a
laptop is not a preferable solution for me; I would probably rather switch
to Linux full time - after all, I would have to run that as my hypervisor,
So, the answer to your question is that for some, now is a good time to
consider virtualization. I just don't think there as a single solution for
everyone or for every installed system.
How to interact with
the users? Do you think that the community is able to generate products?
I think that as in all computing communities, there are spots of talent. We
have some people who are capable of doing this. We have others who are
capable, but who don't feel the need or the desire to get involved. And
finally, we have others who simply do not use the platform to such an extent
as to be in a good position to contribute, other than to talk about what
they like and don't like (and that activity is also necessary).
I do feel that it is important for any project to maintain strong ties to
its audience, listening to changing needs, responding to requests, and
making regular announcements of what is coming and (hopefully) when to
expect the next "thing," without making false or misleading claims. So,
efforts to provide information must be carefully coordinated with those who
know of such things to avoid misstatements and to maintain realistic
practically all computers and notebooks are manufactured by the same factory
today, why support ThinkPad hardware exclusively?
There is no single "standard" in portable machines. As a niche OS, we must
be selective in what hardware we support. Thus, if we *must* choose
something, why not choose something which is familiar to most OS/2 and eCS
As an IT consultant, I *only* support ThinkPads. Like anything else, a good
portion of this is because I know them better than I know other brands. I've
recently worked on a few Samsung Ultrabooks, and I found them to be
sub-standard. So, I have a personal preference for ThinkPads. If someone
would present me with a viable alternative, a standard architecture,
powerful machine, I might change my thinking.
Q: Will your new company create a
central knowledgebase for OS/2 information? Serenity Systems and Mensys
haven't created such a resource.
I don't know. It sounds like a great idea, but our community tends to lack
the motivation for such grand projects, and there is certainly little profit
potential in such a resource, making it a hard sell for a commercial
enterprise. I'd like to see such a thing, but perhaps as a Netlabs sponsored
project, or even a revamp of the venerable
Test the program:
Wanted a nice looking eComStation desktop, UPGRADE it now with NeoWPS.
Comment this article.