Issue #586 : : January 6, 2009
In this issue:
- Spectacular Software Fails; Goes Mainstream
Write the editor. Make him smile!
Through Paypal, consider donating $25 in support of upFront.eZine. Or else...
We're trendy. We have a Weblog. WorldCAD Access.
After 23 years of writing about the computer-aided design business, I have learned that there is nothing for sure -- except maybe AutoCAD.
Entrepreneurs dream up great ideas. The ideas may well be truly great, or sometimes are just great to their eyes alone. Or maybe the "new" ideas are clever/desperate attempts to ride the coattails of some other people's great ideas.
Similarly, public relations people arrive, are friendly for a season, and then go some place else. During their season, they provide me with my greatest amusement when they assure me of a new product's impending spectacular success due to the outstanding management team, tens of millions in funding, and surefire customer demand.
- - -
TechCrunch's Robin Wauters writes about Cuil's spectacular failure, the search engine whose coattail-riding business plan was to out-google Google. It would be a spectacular success, PR people promised the media during pre-launch briefings. But, Mr Wauters writes:
"With the end of the year approaching, I took a peek at how they’re doing traffic-wise out of sheer curiosity. After all, with no less than $33 million in funding and a founding management team consisting of ex-Google search experts, something had to give, right?"
He found that the number of Cuil's unique visitors is close to zero. < www.techcrunch.com/2008/12/27/cuil-fail-traffic-nearly-hits-rock-bottom/ > After 23 years in CAD, I've learned that nothing's for sure -- except maybe AutoCAD.
Spectacular Software Fails; Goes Mainstream
The CAD world has its share of failures as spectacular as Cuil. Here are some that I consider significant. The lesson to be learned is that each one's failure devastated its founders, yet ultimately benefited customers.
1. Xitron XCAD
XCAD was a Windows-based 3D solid modeler programmed in Asia. As I recall from the press conference at A/E/C Systems '94, the company's master plan was to (1) sell the product for $495; and (2) sell one million copies.
The price was shockingly cheap for the time, and the sales goal similarly audacious. We media expressed our skepticism in that press room: One million copies? In return, the company's American marketing folks expressed their shocked at our reaction. In the end, we were right. It didn't sell well, after a couple of years they retreated to Asia, and today I find no current references.
(Don't confuse Xitron's XCAD with the X-CAD name that Alibre used for its pre-release of Design Xpress, or with some variants of IntelliCAD sold as XCAD.)
More than a decade later, however, no one 3D MCAD package has attained the gold record of one million commercial copies. XCAD failed, but their fresh idea of low-cost MCAD software is now mainstream.
2. Display-list Processing
In its early days, CAD programs displayed drawings slowly. How slowly? In late 1985, I timed AutoCAD v2.1x regenerating the famous Nozzle.Dwg on a mathchip-less XT-class computer: 4.5 minutes. How painful? Prior to AutoCAD v2.5, every one of its zooms and pans required a drawing regeneration.
High-end graphics board companies were looking a way for potential customers to justify purchases of their multi-thousand-dollar products. Some figured out how to cache to the display list in the memory on their boards. (I think PixelWorks was the first, Nth Graphics the most heavily marketed, and Artist Graphics/Control Systems the most successful.) Redraws plummeted to under two seconds.
Then a clever programmer figured out how to write a display-list driver that would work with any graphics card. The end came when Autodesk hired away a programmer from Vermont Micrographics, and then implemented display-list processing in AutoCAD's own drivers. (They subsequently lost the inevitable lawsuit by VMI.)
While the display-list market was crashing in slow motion, numerous desperate entrepreneurs tried to promote their replacement drivers as better than AutoCAD's. Maybe their's was better, but who cared when fast came free with AutoCAD. I recall the pair of programmers who visited upFront.eZine with their nicely packaged display-list driver and its wonderfully-designed logo, and I felt bad for their doomed enthusiasm.
All companies selling display-list drivers are gone, but their fresh ideas are now mainstream.
3. Design Variations
It has been nearly a decade since John Lynch visited me with his new idea for 3D software. He formerly held the prestigious title of chief technology officer at Autodesk, but left the firm to pursue a then-radical concept: to make AEC software as easy to use as TriForma, the then-fireworks of a 3D mechanical CAD package whose paradigm-shifting interactivity is now standard in the industry.
Mr Lynch named his software "Design Variations"; if you want to know what it did, take a look at SketchUp. His software had just six commands: profile, extrude, place/delete component, cut, offset, and taper.
I still have the product brochure Mr Lynch left with me. Here are some are the notes I took, describing the capabilities of DV in 1999:
Some of these ideas are only now being implemented in AutoCAD, eight to ten years later. The next time I saw someone from Design Variations, he admitted retail sales were poor, so the plan was to market the APIs. Today, Mr Lynch is with Bentley Systems.
Design Variations failed, but their fresh idea of interactive AEC software is now mainstream.
Q&A With Jim Quanci of Autodesk
[Some former third-party developers for AutoCAD have complained to me that Autodesk is "killing" their business. I asked the head of Autodesk's develop network for his company's side of the story.]
upFront.eZine: A former developer says that Europe used to have three third-party developers writing mech add-on software for AutoCAD. Autodesk bought Genius, which lead to the death of the other two.
Jim Quanci: Autodesk bought two mechanical design oriented partners in Europe: Genius (which led to AutoCAD Mechanical) and another that added several design wizards to Inventor. I believe another CAD vendor purchased another one.
In this changed environment, a few ADN members may have disappeared, though I think you will find a number have taken advantage of the opportunity to change their business. For instance, EMT was one of our larger 2D mechanical design partners in Europe that changed their business and eventually their name. I personally worked closely with them for several years after Autodesk acquired Genius.
upFront.eZine: Some third-party developers tell me they could no longer compete against Autodesk's overwhelming marketing and sales team. As well, they could no longer be members of ADN, I am guessing?
Jim Quanci: When Autodesk acquires an ADN member, their competitors that are also ADN members continue on as ADN members. We do not force them out of the program and we continue to support them. No two companies do 100% the same thing and all partners have varying areas of strength and value-add. We continue to support these partners, what some call 'co-opetition'. This has proven to be a good open approach for Autodesk, many partners, and our joint customers.
On Autodesk acquiring a partner's competitor, partners have taken a few approaches to the changed competitive landscape. Some partners reposition/refocus their business and continue adding value to Autodesk products, which includes:
Note that there are many successful ADN members in the data management and FM [facilities management] markets, despite the fact Autodesk has acquired companies in these two areas.
Recognizing no one acquisition may get Autodesk all that is needed to "democratize" a product worldwide, some partners pursue having Autodesk (or an Autodesk competitor) acquire them too. Witness:
Net-net, the world is a big place and customers needs are highly variable. Thinking any one company can deeply and fully meet most customers needs is just wrong. Customers want and need choice. That's why Autodesk is so committed to wide scale partnering -- both in sales and software. For entrepreneurs, Autodesk's commitment to wide scale partnering creates all kinds of great business opportunities across numerous markets and around the world.
upFront.eZine: What is happening with Mechanical and Mechanical Desktop?
Jim Quanci: AutoCAD Mechanical is a very large, growing and profitable business for Autodesk. We sell lots of standalone AutoCAD Mechanical (not just bundled with Inventor) and customers love it. AutoCAD Mechanical is also quite extensible with partners and customers using it as yet another Autodesk technology platform.
MDT is going away next year as we have finally reached the point where Inventor can do almost everything MDT could -- along with many things MDT could never do (and didn't have the base architecture to ever be able to do). It took quite some time for us to reach this point primarily due to MDT having quite sophisticated surfacing capabilities that have only in the last two years made it into Inventor.
upFront.eZine: I understand that there are a number of reasons why Autodesk would want to promote Inventor over Mechanical and MDT -- more capable in most areas, higher profit for Autodesk, and so on. But that's not necessarily what the customer wants.
Jim Quanci: We don't promote Inventor "over" AutoCAD Mechanical. Many, many of our customers use both. They are different tools and it's about using the best tool for the job at hand. And with pretty good integration, our customers readily mix and match their use of Inventor and AutoCAD Mechanical as they pursue creating competitive advantage through digital prototyping (for real, this is not just marketing speak, talk to our customers).
upFront.eZine: I recall seeing Mechanical demo'ed at AU a number of years ago, and was very impressed what AutoCAD could be made to do through ARX.
Jim Quanci: I suggest you take a look at what AutoCAD has been made to do today with AutoCAD Mechanical, AutoCAD Electrical, AutoCAD Civil 3D, AutoCAD Architecture, AutoCAD Electrical, AutoCAD P&ID and so on.
AutoCAD is a very capable platform that Autodesk and many partners continue to base their products on. Again it's a case of applying the best tool for the job at hand: Revit, Inventor, or an AutoCAD- based product. All are extensible platforms that third parties and customers can tailor to meet their specific needs.
BTW, you are showing your age referring to ARX (ObjectARX) as much of the leading edge application development work with AutoCAD is now done with .NET and WPF.
upFront.eZine: What is your story at Autodesk?
Jim Quanci: I have been working at Autodesk for 19 years, with most of that time helping software developers extend our technologies. It's a great job as I get to see people start and grow their companies. Showing my age.
Long term, it's about the people and not the companies. Many making a good living. Some making a lot of money. A few getting downright rich.
Though some partner companies fail, the people themselves tend to stay in the Autodesk ecosystem and in most part do quite well (financially and personally) whether working for themselves or someone else: paying their mortgages, buying new cars, putting their kids through college and now some are even starting to retire.
The skills they develop(ed) around Autodesk technologies continues to be valued and in demand, which cannot be said for a lot of other companies design technologies that went by the wayside over the last 20 years. Many people were fortunate -- and frankly lucky -- to "pick the right horse" in Autodesk to invest much of our scarce (and growing scarcer and shorter) time and learning in.
[Jim Quanci is the director of Autodesk Developer Network at www.autodesk.com/adn ]
DotSoft ports its AutoCAD-Excel software to Revit. XL2CAD for Revit provides an accurate representation of spreadsheets, and maintains links between Revit and the spreadsheet. www.dotsoft.com/revit
Visual Integrity bundles pdf2cad with Scan2CAD Pro for vector-to-vector and raster-to-vector conversion. www.pdf2cad.com
Siemens PLM Software ships Femap v10, its PC-based CAE [computer aided engineering] modeling and finite element analysis software. The new release creates FEA models quickly through new meshing features. www.siemens.com/plm/femap
ZWSOFT announces ZWCAD 2009, its DWG-compatible CAD software. Download from zwcad.org.ip01-web23.net/download/09exeInstaller/2009/ZWCAD2009EngWeb.exe
Cadalog releases SU Podium v1.6 for Google SketchUp 6 and 7 running on XP, Vista, and Mac. The new release takes advantage of dual-and quad-core processor to double and triple Podium’s rendering speed. www.suplugins.com
Kineo CAM ships its collision detector KCD 2.04.6 software library. www.kineocam.com
JETCAM International releases v16.06.00 of JETCAM Expert CADCAM nesting software, with "Tool Teach Facilities" that allows JETCAM Expert Premium systems to learn users' tooling and profiling preferences. www.jetcam.com
And SolidWorks Labs < labs.solidworks.com > offers these new add-ons:
- - -
These are some of the news items that were posted during our Christmas break at the WorldCAD Access blog <worldcadaccess.typepad.com>:
Cyber Sport patents a new mouse: The Orbita mouse sports a scroll wheel on its top. www.orbitamouse.com
Seminars & Conferences
SolidWorks World 2009 is Feb. 8-11 in Orlando FL USA. www.solidworks.com/pages/swworld09/index.html (I'll be attending this event, even though I got food poisoning the last time I ate at that pink and blue hotel.)
25th Coordinate Metrology Systems Conference is in Louisville KY USA between July 20-24. www.cmsc.org
People/Companies on the Move
Siemens PLM Software appoints Paul Vogel as executive vp for global sales and services. Mr Vogel is the company's former senior vp for sales and services for EMEA [Europe, the Middle East and Africa].
Right Hemisphere promotes Todd Caponi to vp of worldwide sales and George Earle to vp of global enterprise services and solutions.
Autodesk plans to acquire ALGOR for $34 million.
BlueCielo ECM Solutions acquires 49% of PONTODOC Solues de Gerenciamento of San Paulo, Brazil. PONTODOC will be renamed "BlueCielo do Brasil Solues de Gerenciamento Ltda."
Siemens PLM Software says they are now training more than one million students annually on PLM software through the use of in-kind software grants at more than 10,000 academic partners worth more than more than US$4 billion annually.
IGE+XAO reports a Q1 increase of 7.1% over a year ago. "This increase is all the more significant as it was registered in the context of a tense economic situation," says Alain Di Crescenzo.
"Tailoring AutoCAD CUI 2009"
Letters to the Editor
Re: IntelliCAD for Linux
"Ben Decker of Caddit had written: 'progeSOFT tried to support Linux with their CADforLinux product. Our conclusion? The Linux desktop community wants a good CAD program. But they want it free.'
"The average CAD user does not want a version of IntelliCAD running under WINE [Windows emmulation]. That's the same problem BricsCAD has had with adoption of their product for Linux. I have worked hard in the past to implement it and came away disappointed.
"I'm waiting for BricsCAD's NATIVE release and our firm will switch if it is as fully functional as the PC version."
The editor replies: "Linux versions can never be identical to Windows versions, because Linux lacks some Windows features, such as OLE. To be sure, Linux is bound to provide advantages over Windows, such as lower cost. During this recession year, the advantage of lower cost may be beat out the advantage of object linking and embedding."
"I was glad to see my thoughts on IntelliCAD and Linux sparked some comments, both with and without reservations. I'd like to address the more reserved views expressed. The computing needs of AEC firms fall, very broadly, into two groups:
"The first, smaller, group can easily run Linux. The one thing preventing the second larger group from running Linux is the lack of CAD.
"What would happen if IntelliCAD supported Linux? The stumbling block would be removed, allowing an AEC firm to switch to a more stable, secure, and economical OS. Given the current economic situation, AutoCAD's price, and Autodesk's forced upgrades and subscriptions, I can't think of any reason, save inertia, a firm would decline.
"Let me highlight two points here. First, most AEC firms I come to know fall squarely into the narrow range Gartner gives for cost-effective Linux migration, namely 'relatively few applications, and these applications are fixed-function' . So, assuming a Linux IntelliCAD, a Linux-based AEC firm is a pragmatic proposition.
"Second, Ben Decker's response concurs with Dave Lorenzo's remark: Linux users are unwilling to pay for Linux CAD. Only, it seems, this is not the pertinent demographic. The target group for a Linux IntelliCAD would be in all likelihood Windows users: those who through choice, replacement needs, or economic pressure decide to migrate. This group, unlike Linux users, is accustomed to paying for software. My guess is they won't fuss if their licensing costs suddenly shrink to only that of IntelliCAD.
"To put this second point in perspective let's look at some numbers. Linux, by whatever metric, accounts for only a tiny portion of the OS market. For instance, Hitsink gives it 0.83% , W3counter 2.11% .
"However, the growth rate for the last year (November 2007 - November 2008) is solid and is about the same when compared with that of Mac OS X:
Mac OS X Linux
"It is very understandable why Linux IntelliCAD is not considered compelling at the moment. At the same time the growth potential seems very large indeed. If, for the reasons outlined in the previous letter, we are witnessing a gradual exodus from Windows , it seems counter-intuitive that in the present economy most people will prefer the Apple premium.
"So, to reiterate the my central point, the ITC should keep an open mind and think ahead. Their priority now is IntelliCAD 7. But do keep on the back burner a simmering pot full of Linux stew."
- - -
Re: 3D Mouse
"Interesting read concerning the 3D mouse (hmmm, I want one!).
"One comment: you said 'isn't it obvious when panning a view that the graphic point under the mouse cursor should stay under the mouse cursor? This is like grabbing somewhere onto the scene and dragging it around. A few programs do this for orthographic views but hardly any do this for perspective views.'
"When you pick a point in an orthographic view you are actually picking a column of points (or to be more exact, the X,Y of the screen and ignoring the Z). As there is no perspective, a pan will still work because everything remains in line.
"With a perspective view, you cannot do that because as you pan the vanishing point changes, thus changing the points on the Z axis (think of the view from a train). On the other hand, you can pick a "point" in a perspective view and rotate on the spot (or keep the camera stationary while rotating it)."
Spin Doctor of the Moment
"Less fossil fuel is used to produce, install, and maintain a Country Estate PVC fence than is used to grow a forest, protect it from insects, harvest and mill the trees into lumber, and paint and maintain a wood fence."
"It was the Wild West. If you were alive, they would give you a loan. Actually, I think if you were dead, they would still give you a loan."
Copyright 2009 by upFront.eZine Publishing, Ltd. All rights reserved worldwide
Article reprint fee US$250.0 and up.
trademarks belong to their respective holders.
"upFront.eZine," "Talking About CAD," and
"the business of CADg" are trademarks of