What was that early spark?

I never used SolidWorks 95, but I evaluated 96, and bought 97. I evaluated against Autodesk Mechanical Desktop, Micro CADAM Helix, and SDRC Artisan series. Mechanical Desktop was a full-on nightmare. Helix didn’t seem to have much momentum. I would have bought Artisan series because of the ability to expand into something bigger (full-blown SDRC Master series, later merged with the Unigraphics/NX product) except that the company that hired me to do the evaluation expected me to determine SolidWorks was the right choice, mainly on the emotional gut feeling of a couple of people. I don’t usually cave in to pressure to do what is expected of me, but this was a new job I wanted to keep. The technical results of the evaluation favored SDRC slightly due to modeling capabilities and “scalability” – SDRC was one of the oldest players in the CAD market, but SolidWorks won in terms of ease of use and price.  We were changing away from HP Solid Designer, the fore runner to CoCreate, the fore runner to PTC Creo/Direct. The products we designed were plastic medical devices, with some organic shape. Solid Designer and ME10 (for drafting) got the job done (ME10 actually got great marks for creating drawings), but there was a lot of wrestling, little expertise in our company and the entire system from HP including a unix workstation and remote support was unbelievably expensive ($10k/yr recurring costs – in 1996).

So SolidWorks won the evaluation on some emotional gut level – something other than strict CAD capability. In 1996, SolidWorks was a breath of fresh air because it worked based on standards – Windows standards. This was a time when many people were new to the whole computing world, so interface was key. The fact that I didn’t have to learn unix was a huge relief. I would have had to teach the rest of the department how to use the OS as well as the software. Techniques you learned in the Microsoft OS, like Ctrl-select, right click, copy/paste, and so on could also be used in SolidWorks. That seemed to have some value to me.

I was given my first assignment before they sent me to training. I trained myself using the help files, and a couple of tutorials that they had back then. My first original model is shown on the right. I started with the bucket, then the lid, then the big beige device, and finally the blue holder. I was working from a physical clay model created by one of the industrial designers. So my first SolidWorks model was marrying a sculpture to a boxy plastic part. I learned all there was to learn about lofts in the course of 3 late nights at work, and traded some big emails with a reseller (CimQuest) who really did their best to teach me the software and report bugs.

After the first project, I was excited to use the SolidWorks software because it didn’t crash as much as Mechanical Desktop, it was easy to learn if you thought geometrically (like a mechanical engineer), the company behind it seemed genuinely interested in helping me, and it worked in a cost effective environment. It’s true that it wasn’t as powerful as other things on the market, but my needs only barely exceeded its capabilities.

Times have changed. Windows interface standards aren’t quite as valuable as they were 15 years ago. Hardware can be pretty cheap. And these days, Windows is in danger of looking like an expensive and hard to use dinosaur, when you think that a pocket sized device could easily be your main compute center, with interface (keyboard and display) being the cumbersome parts. These limitations could realistically be solved with a tiny portable projector, a projected keyboard, using the phone display as a track pad, and/or voice commands (I’ve been using voice commands and text dictation on my Android for a couple of years). Yes, this all exists today. The only thing that doesn’t exist is the ability to install complex programs like CAD directly on a pocket sized device like a phone. Cloud would solve that problem, but there is no reason why you could not adapt CAD to run on Android and iOS (both basically unix derivatives). Current phone CPUs and storage are far more powerful than the desktop units we had in 1996 (2011 smartphones = 1.5 GHz dual core CPU with 64 GB storage, 1996 desktop= ~200 MHz CPU with  3 GB hard drive).

Anyway, all that just to say that times have changed.

So if we assume that there really was something different about the early SolidWorks company and software that attracted engineers and designers, and we also assume that they have lost that “mojo” and don’t have the same appeal that they used to have, how would you go about recreating that? How would you make SolidWorks a desirable destination again? How would you make them lovable once more?

In a lot of ways, it’s a great thing that SolidWorks is rewriting the code for V6. Let’s just hope they are recreating that sense of wonder and freedom that they gave us in the early days. Yeah, maybe I’m romanticizing the “good old days”, but clearly a lot of people don’t feel the same way about SolidWorks today that we used to. It’s not clear to me that the current bureaucracy has that same beauty of vision that the “founding fathers” had. The people in charge now are very clearly business people, not engineers, and I doubt they could form a clear and cohesive statement that would re-capture the imagination of potential users without resorting to marketing babble. SolidWorks used to be sold to users, now it is sold to managers. That spells out the change clearly enough.

I accept the possibility that there is no going back to the “good old days”, or the feeling we had back then anyway. It could be that for better or worse, SolidWorks has “grown up”, and often grown ups simply don’t understand how to have fun. Companies that grow beyond the human scale are very difficult for humans to relate to, so the relations have to happen on the organizational level – very impersonal, cold and lifeless.

Maybe the only answer to breathe life back into SolidWorks is to make a new product like MOI into the next SolidWorks. It’s not clear to me that SolidWorks V6, making relationships at the larger organizational level rather than the personal, will be the next SolidWorks. SolidWorks V6 seems more likely to be the next PTC.

What do you miss about the old SolidWorks?

13 Replies to “What was that early spark?”

  1. Ok this is my distillation of the long Steve Jobs autobiography into a simple sentence:

    Have total focus first on creating the best, most beautiful solution you can to improve what people can do with technology… and only when you’re sure it’s awesome, then start to give a lot of a shit about the marketing, cos the first bit just made your marketing job a lot easier.

  2. This is all so reminiscent of Apple! You need an energetic entrepreneurial designer/engineer/visionary driving the direction of a company like solidworks. The corporate robots should be there to support them and enact the vision… not be the one’s trying to create a vision for something which they don’t truly understand. I wonder how many thousand hours have any of the current board of Solidworks spent designing real products and feeling the pain of using the existing tools to make that happen… errr. Zero hours?

  3. I remember the feeling that “they” (SW) weren’t that far away and when I would talk to support and suggest things, you had this sense that they actually understood your issues. Like they had experienced it themselves. Now it’s more like talking to a distant software developer that has never really experienced what I do on a daily basis.

    I think some of that is the size of the company, which is somewhat of an inevitable problem. That’s not to say that there’s not better ways to deal with the growth. I think it’s very clear that the goals that SolidWorks now has are entirely different than the goals starting out.

    Craig
    SolidImprov.com

  4. According to my quick Google search John McEleney, who went on to join Cloudswitch in 2009 and sold it on to Verizon in 2011, was SW CEO from 2001 until 2007….so it seems likely the off road adventure started later on in his tenure and continued under Jeff Ray who was seen at the time as something of a DS caretaker, being a business guy rather than technical, and then came increasingly under the thumb of DS. I remember John because I gave him a hard time on the SW forum when he hid in his office and said nothing while we waited 6 months for SW2005sp1 to come out….never did apologise to users or offer an explanation for the delay… and that would have been early-mid 2005?…I actually didn’t buy SW again for 3-4 years – it was the final straw in the stuffed up sp saga for me. As I recollect all SW2004 sp had multiple reissues.
    Its only hazy now but I seem to remember people thought he was losing his way a bit as a leader… vaguely I remember the circumstance as being he was half removed and half left to care of a sick relative or something.
    I think it was about when those automated checking/design apps appeared…there appeared to be a bit of a crisis about what to do next development wise and how to address the quality issues.
    I remember too Matt was critical of those apps. Funny how that now seems trivial in comparison to our present complaints.
    If we consider SW have been experimenting and working toward the cloud for 5 years or more I guess that puts the beginning of what turned out to be decay about 2006 as well. I suppose the implication is that McEleney either seeded the wrong direction before he left picking the cloud was the next big thing or it was something DS imposed on them which has gone from seeming like a viable idea/solution to a disaster the further along it has gone under their direction. Maybe its a mixture of factors only insiders can ever know. Somehow though I don’t think Jeff Ray masterminded the cloud and killed SW.
    Possibly then its not entirely fair to say DS set out to capture their customers but that’s how various strands of enterprise, competition, choice, uncertainty and happenstance have woven together. You rather suspect things just weren’t thought through on a number of levels and critically no one really anticipated the extent of user objection.
    Feel free to add your own interpretation. 🙂

  5. @Ray Regan
    Ray, for me, I really thought the software was improving up to 2007. Crash bugs were way down, but little bugs were significantly up.

    2008 is when the direction really started to bother me. They made a lot of changes to the interface, and the changes weren’t so much what bothered me, but the fact that there were fewer options, and the changes were set up as defaults when you opened the software. From a CAD admin point of view, it was awful. I got the impression at that point that SolidWorks considered the software as a showcase for egos rather than a tool that customers bought and used. It took 2 releases to fix what 2008 broke.

    In that time they were using some wacky research methods to figure out what customers wanted. The PhD researcher leading the wacky stuff doesn’t seem to work for them any more. It was strange stuff, and lots of people were really annoyed by questionaires with randomized choices, where none of the choices were good. There was some huge explanation about how the unwashed could not possibly understand the wisdom of the method, but most people saw it as utter BS. It wasn’t the source of the problem, just another symptom.

    Before 2008, we were mostly concerned about bugs and crashes. Then in 2008 we were concerned about continuity of the interface, administration and options. After 2010 many are concerned about just having a tool. So SolidWorks has progressively created building anxiety. The stuff we are concerned about now is far more fundamental than it used to be. I mean I used to complain about picky stuff. Now I just wonder if I’ll be able to use the software at all.

    The underlying cause I would attribute to corporate growth, hiring a non-engineer CEO, and most of all, Dassault meddling with something they didn’t understand.

  6. I’ll take a guess that it was late 2004 or early 2005 when the decline had its roots which allows for the existance of the last reasonably focused release which was SW2007. Who was CEO of SW then? Was there a change of CEO or some other goings on off the ball at that time?…don’t remember… Of course the further the story goes along the faster things have declined….

  7. Matt, what year, what version, what ownership, did it start to go wrong?…. start to step away from doing what the customer wants, and putting their lean toward marketing SolidWorks? i.e. gettting more users signed up. I bet it was those 2D AutoCad users that enticed them. This mess must be their fault! 2D, 2D, 2D! Those Bas…ds!

  8. I do agree with your statement here:”The people in charge now are very clearly business people, not engineers,”. Once the business people being the CEO of a technology company, he will start to cut cost, play political games do some fancy stuff which is not so technical, and then you will see the product development slow down and compare with other CAD package, SW no longer competitive.

  9. If geometry creation really IS a done deal, decaf processors should be able to work in reduced level of detail but still be able to tell you with certainty if your parameters and geometry will resolve, because they comply with ‘the rules’, instead of resolving on-the-fly. Unless you want to distribute over network or cloud as you go. Autodesk’s 123D direct edit hobby app, 2GHz/2Gb minimum spec recommended processor/RAM, will run on a windows tablet with 1GHz processor/2Gb. It’s slower, naturally, but usable if the entity count isn’t too optimistic.

  10. typically the proof of innovation comes from a patent. I did a google patents search on dassault and didn’t really see anything new.
    https://www.google.com/search?tbo=p&tbm=pts&hl=en&q=inassignee:dassault&num=10#q=inassignee:dassault&hl=en&tbm=pts&source=lnt&tbs=sbd:1&sa=X&ei=NNABT4HuKcPg2QWa26WaAg&ved=0CBsQpwUoAQ&bav=on.2,or.r_gc.r_pw.,cf.osb&fp=9cdf1dc355965aeb&biw=1280&bih=680
    solidworks hasn’t filed one since 2009.
    https://www.google.com/search?tbo=p&tbm=pts&hl=en&q=inassignee:solidworks&num=10#q=inassignee:solidworks&hl=en&tbm=pts&source=lnt&tbs=sbd:1&sa=X&ei=BNEBT-CbMbSlsALrmqWFCQ&ved=0CBsQpwUoAQ&bav=on.2,or.r_gc.r_pw.,cf.osb&fp=9cdf1dc355965aeb&biw=1280&bih=680
    I would think if the spark is going to be brought back it would need to come in the form of innovation.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.