RE: [SLUG] Re: OT - CAD vendors advising customers to outsource--outsourcing works, when limited

From: Bryan J. Smith (b.j.smith@ieee.org)
Date: Sun Oct 31 2004 - 12:59:28 EST


On Fri, 2004-10-29 at 17:49, Ken Elliott wrote:
> Fighting it is *exactly* what we should do.
> Not by passing laws, of dumb things like that.
> But through increased productivity.
> We have to earn our higher pay.
> To say we can't compete means we might as well roll over.

The problem is you are fighting a 3rd world country.
American cannot win because the playing field is not fair.

American workers have been the most productive per unit cost in the G8
since the early '80s. Despite popular thought, Japanese labor costs
passed American workers then. American companies also started adopting
Just-in-Time (JIT) and other manufacturing and logistical approaches,
concepts actually invented in the '50s by US engineers, but adopted by
the Japanese in the '70s. America learned, adopted and overcame,
largely because American engineering institutions _did_ start adopting
their curriculum almost immediately based on Japanese success in the
'70s.

The result is that Japan basically dove into a depression in the '90s.
As their cost of living rose dramatically, and their productivity per
unit cost sank, the economy did too. And, BTW, America owns much of
Japan (not vice-versa), just like the British own much of America.

Unfortunately, now we have a different problem today.

First off, China. China does not respect IP, and piracy of engineering
designs are rampant. So is forgery of American regulator and industry
certifications. Although the G8 is using this as a "bargaining chip" to
the entrance of China into a new "G9." Environmental issues are also at
the forefront.

Secondly is the problem of outsourcing to not just not G8 countries, not
just 2nd world countries, but 3rd world countries. India is slowly but
surely increasing its standard and, therefore, cost of living. It is
now considered "too expensive" compared to fledging software industries
in Ireland, Israel and several others. In any case, we're talking 1/4th
to 1/20th the cost of US manpower. It's not like a G8 nation where
we're talking percentage difference, but _several_times_ over.

Fortunately, the problem is that software is a 100% _social_ construct.
>From the highest-level GUI interfaces to assembler (which, despite
popular software developer assumption, is _nothing_ like actual
execution in silicon). Because it is social, it is affected by locale.
And it is because of locale, things like support, user-interfaces and
most other details can_not_ be outsourced with good results.

Reality? Business leaders _must_ understand what can and cannot be
outsourced. I've written requirements for outsourced software. You
have to "baby" them. Outsourced software development _must_ be limited
to "core, geeky" code development. Anything that remotely presents
itself to the user, the American user, must be written and/or integrated
by American developers. Anything that requires business models or other
social features that are American locale, must be written and/or
integrated by American developers.

Problem? American business leaders and the companies they direct
continue to be dominated by 3 year or less profit model and planning.
With fewer and fewer traditional engineers being graduated each year,
and over 50% of them being foreign born and returning to their foreign
lands after graduation, this is only getting worse. I may be biased,
but traditional engineers who grow into managers grow into the best
leaders of businesses. Unfortunately, there are so few these days,
there is a massive void of _sound_ business planning beyond 3 years.

Which is why the problem is so many variables mashed together.

> Civil engineering on Pro/E? That's pretty funny.

Stress, modeling, HVAC, etc... these are _key_ civil engineering
concepts. Just because most civil engineers focus on statics instead of
dynamic doesn't mean their isn't a place for the latter in the
discipline. The industry has just decided to focus on

> Pro/E is strong in machine design only. That's what it's designed for.

By the same viewpoint, you would argue that DTP is only for publication
then, correct?

I would argue that Word Processing is a pre-GUI technology eclipsed by
DTP with the proliferation of the GUI.

> AutoCAD is a joke in that field, compared to Pro/E - Agreed. But Pro/E
> would be a joke to a Civil engineer, Architect, and for GIS.

Again, I would argue that 2D layout is a pre-3D technology eclipsed by
3D engineering frameworks with the proliferation of the 3D GUI.

> AutoCAD can be a platform for all of these, and that's why it's like a
> Swiss Army Knife.

If Pro/E, among other systems, were as popular, it too could have
everything from COGO to GIS to TRAN. Don't confuse popularity and
availability of add-ons with _inherent_ capability.

> But if you compare AutoCAD to ANY specialized app, it loses.

Again, you make the assumption that Pro/E is "specialized" from your
viewpoint. From my viewpoint, AutoCAD is _limited_.

Several people are making the same statements about Sun's Looking Glass,
it's "specialized 3D," when Looking Glass is one of the most _general_
_purpose_ frameworks that anyone has _ever_ written for GUI development.

Assumptions, familiarity, lack of attention to the inherent capability,
these are why such statements are made when they are simply not true.

> Just like a Swiss Army knife is not as good as a screwdriver.

But you assume Pro/E is a screwdriver when, in fact, it's a fully loaded
racket set with cutting and refining ends as well.

> But that doesn't mean AutoCAD sucks, it means it's a different type of
> tool.

Yes, it _is_ a _different_ type of tool. But it is not any more
"general purpose" than Pro/E, it is just more _proliferated_ and
_common_ than Pro/E. Therefore, it has more add-ons.

But it not any more "general purpose."

> I did CAD/CAM/GIS consulting for 15 years. Every product has its
> place.

I have spent almost 2 decades in the civil/environmental,
mechanical/aerospace _and_ electrical/computer fields -- _all_three_ of
the core disciplines of engineering. The term "general purpose" changes
with that viewpoint.

> Example" Compare Pro/E to Rhino for car body design. Rhino blows it away.
> Rhino is fantastic - like modeling clay on a PC. Pro/E simply can't do that
> class of surfaces. BTW, Rhino is $795, easy to learn, and runs on a PC.
> Pro/E is several thousands and takes months to become productive.

Other people would say the same about Windows v. UNIX/Linux.

Again, assumptions, familiarity, lack of attention to the inherent
capability, these are why such statements are made when they are simply
not true.

> I disagree. They've been losing market share for the last 5 or 6 years. But
> they are starting to turn it around. They only lost $98 million last year,
> and actually show a profit for this year.
> http://www.caddigest.com/subjects/pro_engineer/select/011602_cadreport.htm

Show me software companies that are not losing money these days,
_unless_ they are #1 in their market. It is not because of AutoCAD, but
because of the economy.

> From what I've seen, and dealers I've known, PTC has a habit of treating
> their dealers badly. Autodesk has done the same, and worse. I dislike
> companies the screw over my friends.

Autodesk uses an "oneback" document compatibility strategy like
Microsoft. I call this one of the two drivers of "Hostageware."

Bentley Systems has always maintained document compatibility better than
Autodesk in their products. But is Microstation less popular than
AutoCAD

> Example:
> http://www.caddigest.com/subjects/pro_engineer/select/grabowski_rand_sues.ht
> m
> No, Visio _made_ IntelliCAD a competitor.
> Softdesk started development of IntelliCAD, but Autodesk bought them.
> IntelliCAD was sold by Autodesk after the Justice Department forced them,
> and then Visio purchased IntelliCAD, and
> finished developing it. There was quite a fight between Visio and Autodesk
> for a while. After MS purchased Visio, MS pushed the product into their
> "shared source" experiment.
> I don't recall any non-Windows ports.

AutoCAD had several until the Autodesk-MS alliance.

Anyhoo, same result. Goes to show that legislation and regulation are
rather incapable of making any difference.

Which is why I am pushing for "peer professional" self-regulation. One
of the best mechanisms we have is the engineering discipline.

Unfortunately, the professional engineering organizations and state
boards of professional engineers seem to be a bunch of narrow-minded,
bridge-build civil engineers who don't think software engineering is a
real discipline with different levels of engineering, engineering
technology and technicians. Hence nothing gets resolved.

On the other flip of the coin, the IT industry calls themselves
engineers with little regard for what the discipline actually entails.
No, it's not "the degree" BS, but the _actual_ principles and
application of risk analysis, microeconomics and attention to ethics.
Whether or not you get those from a degree and 4 years of experience or
12-16 years of experience doesn't matter, it is still an _involved_,
peer-based education.

Knowing the specifics of hardware and networking equipment and how to
plug them together is _not_ engineering. That is technology.

We really need a peer infrastructure to address _serious_deficiencies_
in ethical application of technology. From ATMs to E-voting to
financial networks to power plants -- I tire of seeing horrendous
decisions made.

ATMs and E-Voting that run Windows not because Diebold wants to, but
because consumers want it do. Engineering control systems that are
remotely accessed by IT professionals from Windows only to be infected
with viruses, not because the engineers allowed it, but because they
were overriden by IT and business managers. Financial networks that are
no longer separate from general networks for cost reasons, not matter
how much professionals try to persuade them otherwise.

Legislation and regulation will solve nothing. God knows I've seen the
best designs get "torn down" because departments like the Office of
Thrift Supervision (OTS) don't care about good designs, they want to see
consistency, so everything is implemented as the "least secure" system
in a company. Just makes me want to pull my hair out. The BoPEs in
their arrogance versus the IT world in its gross ignorance. No one wins
and the public trust and safety is not protected.

> I'm not here to defend Autodesk - far from it. I knew the founders, but I
> don't like what the company turned into, and how they put over 90% of their
> hard working dealers out of business (from over 3,000 to about 150). But
> PTC has also been a stinker, and a big one. In the industry, it was well
> known that PTC would try to take any order direct if it involved 3 seats or
> more of Pro/E. That sucked. SolidWorks (when I was a dealer) never took
> any order away from a dealer. You were treated with respect while Autodesk
> and PTC would crap on you. Most of the dealers I knew dumped Pro/E, and
> switched to SolidWorks. That's when PTC started hurting.

I've worked with PTC and I really don't know how you can't compare them
to what Autodesk does.

-- 
Bryan J. Smith                                  b.j.smith@ieee.org 
------------------------------------------------------------------ 
"Communities don't have rights. Only individuals in the community
 have rights. ... That idea of community rights is firmly rooted
 in the 'Communist Manifesto.'" -- Michael Badnarik

----------------------------------------------------------------------- This list is provided as an unmoderated internet service by Networked Knowledge Systems (NKS). Views and opinions expressed in messages posted are those of the author and do not necessarily reflect the official policy or position of NKS or any of its employees.



This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 20:15:20 EDT