Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
(ASCEND) "Free" Software vs. Licensed (was "analyse RADIUS logs")
Bob Tanner wrote:
>> > Since I am an ISP we are almost totally Unix. Where is the Unix
>> > software support?
and Dale E. Reed Jr. replied:
>> There are two REALLY different markets. Most UNIX people are [sic]
>> willing to put out money for software. They want their LINUX
>> for nothing and their tools for free. How can Ascend make
>> money there? Now on the NT market, people are willing to put
>> money into products, so there IS a viable market for selling
>> products on windoze.
>> We sell only WindowsNT products (clients run on Win95) and
>> have a lot of UNIX people coming over to our products because
>> its basically turn-key supported products they don't have to
>> compile and muck around with.
then, Julian Cowley chimed in:
>Ok, I'll go and forget that I ever had Netscape or Acrobat Reader (or even
>Livingston's pminstall) on my SGIs, Suns and Linuxen.... How can Ascend
>make money there?
Well, it is obvious that Dale makes money selling software
which would not even exist if not for the original (public
domain) Radius. Dale should be forgiven for having a vested
interest in promoting the one platform for which he has a
product. Much like Ascend's "supported version" of Radius,
Dale's Radius most likely:
a) Costs money (one would presume that the customer
is paying for the "added value", not the freely
available baseline Radius code, but this is not
known to me)
b) Is supported by the vendor (as opposed to being
supported by the usual consortium of users and
contributors to a gnu or other public domain
work)
For some customers, this appears to be an advantage.
The contention made by Dale - that "Unix" means "Linux", and
all Unix software must be "free" to have "market share" confuses
many issues:
1) Unix has been very very good to many people, myself
included. So (was) MVS. So has Windows in all
flavors, but less-so given that it is rather hard
to sell a $200K application to run on a $5K engine.
Rather large checks still come in the mail every
month from happy customers who paid (and are still
paying) significant license fees for significant
software that runs under Unix. Don't try to claim
that people who run Unix are not willing to shell out
serious piles of small green pieces of paper for tools
and software. It is simply untrue.
2) While it is clear that the internet is overwhelmingly
Unix-based, this is not because the software is free.
It is because things work this way. In the hands of an
experienced professional, things work first time and
every time.
3) While public domain applications, combined with
built-in Unix facilities dominate the market for
internet infrastructure, this is NOT because they
are "free". It is because they are BETTER SUPPORTED
than the commercial products. How so? Simple.
3a) Product selection is made less problematic,
based upon the theory that a hundred (or
a thousand) of your fellow professionals
simply cannot all be wrong. Any known
bugs or problems can be reviewed up front
by checking the mass of FAQs, mailing list
archives, and so on.
3b) Product support is made pain-free, since one
can bet that someone else has run into the
same question/problem that you have run into,
and you can follow the adage "Ask, And Ye
Shall Receive". It has always been thus,
it will always be thus, both now and forever.
Commercialization of the net is a recent fad,
and this too will pass. The net was built via
standards and cooperation. If the interlopers
think that they can "get rich quick" think that
they know better, good for them. I for one
have no intention of ignoring a fellow
professional's request for assistance, since I
(like everyone) have benefited much more than
I have contributed.
3c) Public domain software comes with SOURCE CODE.
One is not a prisoner of a vendor's revision
cycles and mixed agendas. If you want to add
or modify, you can. This goes right down from
the application to the Unix kernel, since even
commercial vendors of Unix DO have the ability
to license the source code, and have had this
ability since day one.
4) There are platform alternatives. "Linux on Pentium boxes"
is a very recent experiment in the grand scheme of things,
so the jury is still out. While there are many small
operations who think that a pair of Linux boxes are all
they need to get into the ISP business, many of them at
least "graduate" to SCO Unix at some point. Small
shoestring operations are never a good market for a
software vendor, since they simply lack capital. It is
true that this small segment of the market appears large,
since this group is the most vocal when crying for help
on the mailing lists and in the newsgroups. "Linux for
nothing, and tools for free" is a cute pun on the song,
but much like the original song, it is the anthem of a
"flash-in-the-pan" group.
5) Attempts have been made to create internet servers on
Windows NT. While this may seem like a great idea,
one would hope that at least Microsoft Network would
have put its network where its mouth is, and use
Microsoft's product. Humorously, they only recently
stopped using Unix (in at least the most obvious places)
and started using NT at all. They still do not use the
product that Microsoft wishes to market to other ISPs.
Web Week, July 21, 1997 - Page 26:
"Curiously, MSN doesn't use the Microsoft commercial
Internet System (MCIS), a suite of communications
servers marketed to ISPs, which Microsoft has said
is based upon the server software developed for MSN."
I for one, am not going to trust a product that cannot
even be sold to a captive customer. If MSN won't use
Microsoft's product, why should we?
6) There is a great push from recent arrivals, wannabes,
and "suits" with no sense of history or perspective
to try to make running an internet site something
that can be done by inexperienced/untrained personnel.
This is not because there is a lack of trained
professionals, but only because the "suits" do not
want to pay $150K a year to some "long-haired, drug
crazed hippy" who happens to be very very very good
at what he/she does. What these folks fail to understand
is that the internet is very much a moving target, and
moving targets require skilled sharpshooters and snipers.
Good sharpshooters and snipers cost good money, and demand
highly customized handcrafted weapons.
The "solution" being marketed to these folks is a mix of
GUI front-ends and canned scripts. What these
customers do not realize is that these "simplified"
products simply cannot solve the problem, since the
"problem" is the need for a basic understanding of a
wide range of areas. Therefore, the customers for
these products (perhaps) can handle the day-to-day
repetitive tasks (which are likely handled by cron jobs
on a system run by competent personnel) with these
GUI-based systems, but have no idea what is going on
(or what is NOT going on) when things go wrong.
The result is that the amount of money paid is more than
it would have cost to employ a competent and experienced
professional. The equation is:
( Canned_Apps_Cost + Downtime_Cost
+ Learning_Curve_Cost
+ Lost_Opportunity_Cost
+ Cost_Of_Low_Wage_Workers )
is much much greater than
( Cost_Of_High_Wage_Workers )
If there is even a small group of people moving specific
traditional Unix-based tools (such as radius) to NT, I
think that this is likely limited to specific single-purpose
workstations/severs, and does not imply that anyone is
scrapping Sun Ultras, Dec Alphas, SGI boxes, or any of the
other usual platforms. I would guess that "offloading" a
specific task that is not a part of the basic
mission-critical dataflow to a Win95 or WinNT box is more
of a personal choice.
7) Just to make things worse, many of the default choices
for a professional internet site are simply not available
for (not ported to) platforms like NT, because the
community of developers/users see no reason to take a
good tool and expend effort to cram it into a poor
platform. Therefore, people who select platforms like
NT are FORCED to buy commercial apps, since they have no
other choice.
As for Dale's question
"How can Ascend make money there?" (selling to people
who use Unix")
This is simple. If Ascend's products are not fully
compatible with the industry standards (Unix and
associated tools), people will SCRAP their Ascends,
and buy something else that is compatible. Ascend
sells hardware. They live and die by their market
share, so they are well advised to "bundle" software
tools that reduce their cost of tech support, and
freely distribute high-value software that keeps their
products easy to use.
If Ascend decided that giving away K56 hardware (in
order to maintain market share) was a good idea,
I would expect that I will continue to see free
upgrades to microcode, and at least the occasional
software tool, simply to make sure that I do not
dump Ascend for another vendor.
As for Ascend's ability to sell me software, all
I can say is that FIRST, I need bug-free, organized,
well documented, and bulletproof upgrades to microcode,
THEN, I will think about buying my tools from them.
Hey - I charge serious bucks for "wisdom" like the above, and
people willingly pay it. But here, I offer it freely to my
fellow professionals. Ain't the net neat?
Rather than "carpe diem", on the Internet, once is wise to "carpe PM"
james fischer jfischer@supercollider.com
++ Ascend Users Mailing List ++
To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd: <http://www.nealis.net/ascend/faq>
Follow-Ups: