Executor 2.0/executorv2.0.iso/pc/dos/extra/docs/maillist/text/archive.95/text4345.txt

Internet Message Format | 1996-03-31 | 11.3 KB

Received: from sloth.swcp.com (sloth.swcp.com [198.59.115.25]) by nacm.com (8.6.10/8.6.9) with ESMTP id QAA07683 for <[e-mail]>; Tue, 29 Aug 1995 16:19:40 -0700
Received: from iclone.UUCP (uucp@localhost) by sloth.swcp.com (8.6.9/8.6.9) with UUCP id RAA08910; Tue, 29 Aug 1995 17:19:32 -0600
Received: from beaut.ardi.com by mailhost  with smtp
    (nextstep Smail3.1.29.0 #11) id m0snZqv-000Yc3C; Tue, 29 Aug 95 17:13 MDT
Received: by beaut.ardi.com (linux Smail3.1.28.1 #5)
    id m0snZqv-00005nC; Tue, 29 Aug 95 17:13 MDT
Message-Id: <[e-mail]>
Date: Tue, 29 Aug 95 17:13 MDT
From: [e-mail] (Clifford Thomas Matthews)
To: [e-mail]
Cc: [e-mail]
Subject: Re: Win95, and Why...?
In-Reply-To: <[e-mail]>
References: <[e-mail]>
Sender: [e-mail]
Precedence: bulk

NOTE: I am answering this piece of e-mail before I finish my other
e-mail, so if you have sent e-mail to me or questions (which is routed
to me temporarily) and are awaiting a reply, I haven't forgotten -- I
just want to explain ARDI's policy, since I hope it makes sense with a
good explanation.

>>>>> "Jim" == JIM210  <[e-mail]> writes:


    Jim> I run Exec199o5 under Win95 by simply ...

Thanks for explaining how to run Executor under Windows '95.  We have
recently purchased a copy for testing purposes and hope to incorporate
your (and other people's) suggestions and our experience into a new
FAQ entry or two.

    Jim> I will say I get excited when I find one more program that
    Jim> works under Exec, but like most I get frustrated when I get a
    Jim> string of them that don't.

Right.  And it's *doubly* frustrating that Executor doesn't really
give you a hint as to why it has died, and even when it does give a
hint, the hint is incredibly cryptic.  Two obvious solutions come to
mind: make Executor run a lot more applications and also totally
revamp our error handling procedures.

    Jim> But my main concern about the ultimate success of Executor is
    Jim> not files but it's price ie, what you get for the price.  Up
    Jim> until Exec199m it was a curiosity, a toy-and exciting; barely
    Jim> worth $99.

Pricing is a tricky issue, since different classes of users find
different worth in the same product.  Ever since Executor-MSW (the
first commercial release of Executor was for NEXTSTEP and the only
program that guaranteed to run back then was Word), Executor has been
able to read and write Macintosh formatted floppies.  This ability
alone is useful to some people more than others.  Mac-In-Dos is a
commercial product that can do this and little more.  Still, your
point was understood by us and when Executor/DOS 1.0 was released our
very own packaging said:

    "We recommend purchasing this *first* version of Executor/DOS
     ONLY if:  you want your PC to be able to read and write
     Macintosh formatted 1.4Mb floppies or You want to explore
     the ``bleeding'' edge of Macintosh Emulation or You want to
     be able to purchase Executor/DOS 2.0 at a discount when it is
     released or At $99 the combination of the above reasons is
     >>irresistible<<."

In fact, since 2.0 has taken so long to come out and has been stripped
of a feature or two, we've decided to give the Executor 1.x users a
*free* upgrade to 2.0.  I point this out because it shows that we try
to be blatantly honest in our promotional materials and that we care
about our early adopters.

    Jim> But if Exec2.00 is little different from Exec199o5

There are many things that will change significantly between 1.99o5
and 2.0.  In addition to much more intelligent error messages, we
really should have many other applications running.  It's hard to see
all the progress that we made between 1.99o and 1.99o3 (the first
post-hackathon release), because we did a major rewrite of our memory
manager and managed to introduce a few bugs that have caused many
programs that used to work to fail and because our System 7 filesystem
code is incomplete.

Once we clean up a few penetrating bugs, many more applications will
run.  However, right now we're not working on penetrating bugs, per-se
-- we're working on the NEXTSTEP port, adding error messages and DOS
extender robustness issues.  We've made progress in each of these
areas and hope to be able to get back to improving core compatibility
soon.

Furthermore, it looks like Executor 2.0 will be released as a CD-ROM
with at least a couple hundred megabytes of applications that are
fairly thoroughly tested and known to work.

    Jim> it will be a hard sell at $249 to the public,

The suggested retail price will be $249.  We're already working with a
major mail order house and expect them to offer it at a pretty large
discount, so the street price will still be over $99, but *probably*
will be under $150.

    Jim> but not to institutional/educational organizations *if* it
    Jim> fully supports at least one popular Mac word-proccessor/DTP
    Jim> and those programs they use the most.

Most of the popular word-processors and DTP applications are already
available for DOS/Windows, but we do expect institutional/educational
organizations to purchase Executor if we can run second and third tier
(first tier applications are those made by the biggest software
manufacturers and as such are usually available on both platforms now)
that they require.  One example of such a program is NIH-Image -- a
public domain image processing program available for the Macintosh but
not for the PC.  I'd guess about 30% of our sales in the last month or
so have been to people who want to run NIH-Image.  The point here is
that we have much more information about what people want than may be
obvious.  We talk to our current customers and to our potential
customers.  We're fairly aware of the business issues involved and try
to do the right thing.  Many times when we do something that may seem
incorrect to an Executor user/enthusiast, it may be due to incomplete
knowledge of the Executor user.  This isn't to say we don't make
mistakes -- we do, although hopefully we learn from them.

    Jim> Even then...  But as for the public (ie me included), without
    Jim> sound, Quicktime, and modem(serial) support I would suggest
    Jim> they would think $249 gives you very little in todays world
    Jim> of CD-ROMs(gaming and videos), Faxing, and the Internet.

Setting software prices is tricky -- sort of like selling seats on an
airplane.  One pricing methodology is to initially charge a lot for a
piece of software and then drop the price later.  The people who
REALLY want the software (or who have large software budgets) from day
one pay the higher price, and others wait and then later pay the
smaller price.  I am not saying we're going to do that with Executor,
in fact, we're largely doing the opposite, at least with our pre-Beta
prices, but the point here is that the price set on software is not
simply a function of what will bring you the most customers initially,
it's not even a function of what will bring you the most profit in
your first n months, since pricing strategies have long term
consequences (if you price software too low, some people will dismiss
it solely for that reason).

I have tried to study the pricing policies of other companies and
watch the prices of other software.  I try to temper my desire for
profits (I've been up front about this from the beginning -- I founded
ARDI to make money) with a desire to treat customers fairly, honestly
and as intelligent people.  When we first sold Executor we even
avoided a practice that we thought was semi-sleazy: pricing a hundred
dollar product at $99 instead of $100.  In my mind that was only a
ploy to make the software *seem* less expensive than it is.  However,
I eventually found out that in institutions dropping a dollar
sometimes shifts the price from one category to another (i.e. it might
allow some people to buy the product from petty cash, instead of
having to use the purchasing department).  Live and learn.

The bottom line is that we pay a great amount of attention to our
pricing policies and are trying to price Executor to bring in
substantial revenue to ARDI (initially all the money will go back into
R&D and marketting -- we won't be taking profit for a while), but to
also reward our previous customers.  We have very loyal customers and
we appreciate that.

    Jim> But if I may ask, *with time/money in such short supply*,and
    Jim> * from a strictly business viewpoint*, why work on multiple
    Jim> platforms at the same time? EVERYBODY knows Executors success
    Jim> will come from the DOS/PC world not Linux or NEXT(?). It
    Jim> sounds like its more do to personal interest of individual
    Jim> engineers there and not a business decision.

This is a very good question.  Perhaps it should be an entry in the
FAQ, although I suspect more people wonder about this without asking
than actually bring the question up.

All the core development of Executor is done under Linux.  We decided
to release a Linux version because we had many people ask us for it
and we knew we could do so without polishing it much.  However, since
Linux people are more likely to be FTP users and be Usenet readers, we
have found that for the last month or two our new customers are
approximately 40% Linux users and 60% DOS uses (no new NEXTSTEP
version translates into no new NEXTSTEP customers).  Different
companies use different development tools, and I'm not trying to tell
anyone else how to develop software, but I honestly don't think that
we could have done as much as we did without the Free Software
Foundation tools that we use under Linux (and we used under NEXTSTEP
back when we were developing under it).

Supporting NEXTSTEP is indeed a drain of resources, *BUT* before the
NEXTSTEP market had a sudden metabolic adjustment, we had many
NEXTSTEP customers (they currently make up about 38% of our customer
base) and we greatly value our customers and do not want to dump them
unceremoniously.  As it is, they've constantly had to take a back
seat, since a new NEXTSTEP port would never happen if we were to go
bankrupt.  Deciding how much time to work on NEXTSTEP is tricky, since
it will be easier to do more NEXTSTEP work after 2.0 ships, but on the
other hand, once a certain amount of work is done on the port, we will
suddenly have many more intelligent customers providing us with words
of encouragement and cogent criticism.

    Jim> And when you are
    Jim> way overdue on a project I would think you would do only what
    Jim> is essential and would contribute most to its/your success.
    Jim> Linux?! Next?!  Criticism?No.Puzzled?Yes.

I wish I had more time to go into all the issues you've raised in
greater detail.  I am glad you and others are concerned enough about
us or our product to ask these questions and I think the best way to
reward people who ask valuable questions is to give valuable answers.
However, time spent answering this letter is time not spent coding or
debugging, so I hope you'll understand.

I do think that 2.0 will be well received.  I also think that
considering what ARDI has done so far, on a shoe-string budget, that
once 2.0 is shipping and we hire a few more engineers, you'll see
Executor improve dramatically over the next year.

    --Cliff
    [e-mail]