Will History Be History?
Of late I have been working 48 hours a day to get Print3D up and running. Now that I have some time to breathe I can finally get back to blogging. I have been reading the recent posts from Roopinder Tara, Matt Lombard and Kenneth Wong on Synchronous Technology and would like offer my $0.02.
Some time ago I took the time to figure out Synchronous Technology and wrote a series about it. Since I had to piece things together and speculate how this techology worked, I was glad when people from Siemens let me know that I had got most of it right. I am done analysing the technology. I know the technology is good. It solves far more problems than it creates. But all said and done, what I am most interested to know is whether the time for this technology has come. This is more of a business question, and less of a technical one.
I believe it's time has come. Here is why. The CAD industry has its limits. CAD vendors are spending too much marketing money in trying to eat into each other's market share. Roopinder rightly mentioned that changing religion may be easier than switching over to a different CAD system. As I see it, the easiest way to increase revenue is to increase the market itself. And you just cannot do that with strict parametric modeling system. You need to dumb it down so that it can be used by people who do not have the patience, skill or IQ to create or figure out a convoluted feature tree.
Look what SketchUp did to the ordinary computer user. People of all sorts of shapes, sizes and ages are populating Google Earth with buildings. People, who till recently used MS Paint to come up with anything creative on a computer.
History will testify that things start out complicated and then get simplified in time. I started out in DOS and first programmed in Turbo C. I cannot tell you how much pain and code was required just to display the mouse cursor (basically a block character) on the screen. Then came Windows and life became much easier. We left Windows to handle the mouse and were only bothered with what our program should do when someone clicked it. My point is that in the days of DOS, you and your code needed to be wired up correctly to be able to come up with any worthwhile program. With Windows, the operating system handled the user interface stuff and you were left with the job of adding functionality to the program, which was why you started writing the program in the first place. So in a way, programming became dumber and many more people started programming. But even though programming became dumber, the programs didn't. These new programmers who had a fraction of patience, skill or IQ as compared to the DOS programmers came up software far superior than their DOS counterparts. To emphasize my point, the reason why everybody cannot be a rocket scientist today is because designing rockets is still a very difficult to do.
If you are sailing a boat you had better know which side the wind is blowing. I believe the smart CAD vendors like SpaceClaim, Siemens and the like know the direction of the wind. The others will watch them and say, "Crap!! That's the direction. OK, now lets play catch up".
Yes, history will be history.
Some time ago I took the time to figure out Synchronous Technology and wrote a series about it. Since I had to piece things together and speculate how this techology worked, I was glad when people from Siemens let me know that I had got most of it right. I am done analysing the technology. I know the technology is good. It solves far more problems than it creates. But all said and done, what I am most interested to know is whether the time for this technology has come. This is more of a business question, and less of a technical one.
I believe it's time has come. Here is why. The CAD industry has its limits. CAD vendors are spending too much marketing money in trying to eat into each other's market share. Roopinder rightly mentioned that changing religion may be easier than switching over to a different CAD system. As I see it, the easiest way to increase revenue is to increase the market itself. And you just cannot do that with strict parametric modeling system. You need to dumb it down so that it can be used by people who do not have the patience, skill or IQ to create or figure out a convoluted feature tree.
Look what SketchUp did to the ordinary computer user. People of all sorts of shapes, sizes and ages are populating Google Earth with buildings. People, who till recently used MS Paint to come up with anything creative on a computer.
History will testify that things start out complicated and then get simplified in time. I started out in DOS and first programmed in Turbo C. I cannot tell you how much pain and code was required just to display the mouse cursor (basically a block character) on the screen. Then came Windows and life became much easier. We left Windows to handle the mouse and were only bothered with what our program should do when someone clicked it. My point is that in the days of DOS, you and your code needed to be wired up correctly to be able to come up with any worthwhile program. With Windows, the operating system handled the user interface stuff and you were left with the job of adding functionality to the program, which was why you started writing the program in the first place. So in a way, programming became dumber and many more people started programming. But even though programming became dumber, the programs didn't. These new programmers who had a fraction of patience, skill or IQ as compared to the DOS programmers came up software far superior than their DOS counterparts. To emphasize my point, the reason why everybody cannot be a rocket scientist today is because designing rockets is still a very difficult to do.
If you are sailing a boat you had better know which side the wind is blowing. I believe the smart CAD vendors like SpaceClaim, Siemens and the like know the direction of the wind. The others will watch them and say, "Crap!! That's the direction. OK, now lets play catch up".
Yes, history will be history.
5 Comments:
"I believe it's time has come. Here is why. The CAD industry has its limits. CAD vendors are spending too much marketing money in trying to eat into each other's market share."
No one is eating into SolidWorks market share in the machining job shop market... they own most of it and what they don't own is steadily increasing despite the fact that history based modeling has no way to quickly and easily edit a non-native solid.
Almost every major CAM vendor either has or is working on a CAM product to run inside of SolidWorks. This is happening because the CAD tools in CAM products like Mastercam, Gibbscam, Surfcam, Featurecam, DP Esprit, Smartcam etc. are so far behind the times that it's hopeless to try and fix them.
Unless vendors like SpaceClaim and Siemens step up to the plate and go head to head with SolidWorks and prove to users what a nightmare history based modeling can often be, this situation won't change. I've laid out what both Siemens and SpaceClaim need to do both on my blog and on WorldCAD Access.
No machining job shop, designer or company that I know of is using Autocad LT, yet somehow SpaceClaim thinks they are going to capture a huge market that so far has totally eluded them?
The real market for now is showing the established market that they have the wrong tool and showing them the right tool to use.
It's much easier to convert a market using the wrong tool than to create a market that really doesn't exist... yet. When that market does exist it will start clambering for better tools. Right now that "market" is close to silent.
Jon Banquer
San Diego, CA
http://jonbanquer.wordpress.com/
By Anonymous, At 2:38 AM, October 22, 2008
Deelip,
In the end, I think it will be the market that will make the decision, not the developers. Microsoft has the monopoly to shove the ribbon down people's throats, but the market retaliated, and now mac shares are roughly equivalent in number to Vista shares.
The CAD market is different. There are a lot of choices not just in brands but modeling paradigms too. Non-history modeling, as impressive as it looks, still has some significant limitations that will prevent it from conquering the entire CAD planet immediately. There may always be a segment of design tasks where history and non-history are roughly equivalent.
We will see a certain type of user switch first - people doing simple prismatic geometry. This could conceivably include a lot of new 2D-3D converts.
It may take several years for non-history solid modeling to break with much force into the domain of Rhino. Consumer product design will likely not be done in Spaceclaim or SynchTech for a while. Looking forward to see what Catia V6 does to this side of the question.
Matt Lombard
By Anonymous, At 2:47 AM, October 22, 2008
Deelip,
In the end, I think it will be the market that will make the decision, not the developers. Microsoft has the monopoly to shove the ribbon down people's throats, but the market retaliated, and now mac shares are roughly equivalent in number to Vista shares.
The CAD market is different. There are a lot of choices not just in brands but modeling paradigms too. Non-history modeling, as impressive as it looks, still has some significant limitations that will prevent it from conquering the entire CAD planet immediately. There may always be a segment of design tasks where history and non-history are roughly equivalent.
We will see a certain type of user switch first - people doing simple prismatic geometry. This could conceivably include a lot of new 2D-3D converts.
It may take several years for non-history solid modeling to break with much force into the domain of Rhino. Consumer product design will likely not be done in Spaceclaim or SynchTech for a while. Looking forward to see what Catia V6 does to this side of the question.
Matt Lombard
By Anonymous, At 2:48 AM, October 22, 2008
"Non-history modeling, as impressive as it looks, still has some significant limitations that will prevent it from conquering the entire CAD planet immediately."
I'm not sure it needs to conquer the entire CAD planet. I often hear this statement about its limitations, though. I'd like for once to see a definite list of things that are best done in Direct Modeling and those that are only possible in History Based modeling.
By Anonymous, At 5:28 PM, October 25, 2008
Deelip,
Direct editing is a neat innovation but if you play with spaceclaim long enough you will see that it is very easy to make a model which cannot be easily edited. Such tools will lead to frustrated users.
I feel history based modeling will be around for a long time. It may evolve to become more scalable and robust. I'll take the example of google chrome (they did not create anything new but they made the browser stable and secure and work with multiple cores of the CPU)
By Anonymous, At 10:54 PM, October 28, 2008
Post a Comment
Subscribe to Post Comments [Atom]
<< Home