SWW09: Bits and Pieces
So many little things happened at SolidWorks World that it’s hard to keep track of them. Writing them here will help me remember, and give you a finer grain detail of what kinds of things happen at World.
One night I got in the elevator and recognized the fellow that was already in there as Stephen Wolfe. Mr. Wolfe has had a number of roles in the CAD publishing world. I started a conversation with him, and we talked for a while onto the 7th floor. I’m not sure he recognized me or that my name would ring any sort of bell. We talked about CAD and non-CAD topics. I think the opinions and insight of people who have been around the industry to see trends come and go are very often instructive.
After all of the hoopla around Spaceclaim and Synchronous technology, I think my early skepticism has turned out to be correct. History systems aren’t going to go away, but they are going to get more direct edit functionality. Direct Edit systems may gain their own customers, who are downstream data users who work mostly from imported data. Real design work doesn’t thrive on direct edit as much as some folks are trying to make out. Some analysts got too carried away with the hype early on. CAD must be an incredibly boring thing to write about, where even the hint of a revolution sends folks into histerics. Mr. Wolfe’s opinions of the future of Synchronous Technology seemed to be somewhat more modest.
Wednesday night, after it was all over, Kim and I went to an early show of Cirque du Soleil. Wow. That was amazing. On our way back we had dinner, and were going to get some ice cream in one of the restaurants in the Dolphin hotel, but stopped to talk to some bloggers hanging out in the bar. While talking to folks there, Jon Hirschtick and Jim Wilkinson came up. Jon has shaved his beard and lost a few pounds. “Whoa! It’s the hit squad!” I said, but Jim came back saying that they thought I was the hit squad. Jim’s a great guy, but we don’t always agree on the direction the software should take. He also gets caught in the middle of stuff sometimes, and doesn’t have complete control over things that affect the software, so he shouldn’t get all the blame… Hirschtick, well, it’s easy to get nostalgic for “the good old days” when you see some of the things coming out of the current administration. As I said before, I think Jon is becoming the public face of this administration because users have difficulty relating to Mr. Ray, but everybody loves Jon.
It turned out Jon and Jim wanted to talk. Jon had a great idea, giving me the opportunity to basically “put up or shut up” (my words, not his). He had presented his idea to Jeff Ray, and it would be aimed at improving surfacing functionality in particular. It would give me the opportunity to say everything I think needs to be said to the people who can make it happen, and give them the opportunity to make it happen. I think it’s a brilliant idea from SW because either way it shuts me up and wins me over, and has the potential to get some real usability built into the surfacing functions. If this goes forward, I will be soliciting great surfacing ideas from readers. Even if it doesn’t go forward, please send in ideas for surfacing functionality.
One of the things that Jon said during this discussion was that great ideas come from individuals, not from committees trying to negotiate a settlement (I’m paraphrasing here). He is undoubtedly correct. This was one of the ideas I wrote about in one of my recent articles on innovation vs trend following. It’s true. Design by committee doesn’t work very well. Products of all kinds should be built on a single philosophy that makes it all make sense. SolidWorks is such a huge piece of software, it is difficult to make that actually happen, and the older and more complex it gets, the more it diverges from the idea. Maybe sections of the software could be run this way.
Part of SolidWorks World is certainly the party aspect. Partying is fun. Alcohol usually loosens you up a little, but sometimes it makes people do stupid things. You’re not always careful what you say when you’re drinking. Talking to people when you’re drinking is as dangerous as … well … as making comments on a blog which can be misconstrued so easily. So, short lesson here, don’t comment on a blog without being careful about what you have to say, especially if you have a lot to lose, and second, don’t try to have serious conversations when you’ve had too much to drink. It’s a bad idea, and it can have consequences. You might think you’re in control, but people around you, especially those who have not been drinking, can tell from your glossy eyes and slurred speech that you’re in no condition to talk seriously.
That’s just a paragraph for thought. Not necessarily connected with anything or anyone.
After talking to Jon and Jim, and then getting back to the blogger group, Kim and I were finally going to get that ice cream, but Rich Welch stopped me. Jon and Jim intervened in what turned into a very tense and unnecessary situation. Anyway, we eventually agreed to meet again the next morning. The morning meeting gave us the opportunity to set it all back to zero, but we didn’t make any headway beyond that. He did acknowledge that the people who commented on this blog (about subscription issues) were genuinely upset about something, but it was unclear what he expected from you, the commenters, or from me, or what he had to offer in return. I will do a future post that addresses what I think he needs to do, although I think we have all covered that pretty well in posts and comments several months ago. It is clear to me that none of the suggestions we had are even going to be considered. Their idea of “more value for subscription” seems to be giving you a free shot at certification, and giving away toys at CSWP events. There was some mention at one point about helping resellers to give more technical value, but it was a fleeting mention, without detail.
Kim and I did finally get that ice cream.
I heard 4 different people take credit for the Brainstorm idea. It’s clear that people like it, and that the concept is going to be refined and used more widely. Corporate politics are insane. SolidWorks has squelched this idea from users for YEARS, and now a relatively new employee slips it through, and everyone up the ladder or anywhere in the vicinity is taking credit for it. It’s clear that the reason it was not used before was out of concern about the competition cashing in on the results. Customer driven or Competitor driven?
Charles Culp was one of the more interesting new people I ran into at SWW. He is of course prolific on the SW forums, but did you know he has only been using SW since 2006? Charles has incredible depth of knowledge for only being a user for that short amount of time. We got to talk at the CSWP event. I imagine this is a guy we will hear more from in the future.
Alex Ruiz (aka SWGeek) was another person who was new to SWWorld. Alex is a good natured and ambitious user deeply involved in the social networking scene around SolidWorks. Knowledgable about the software and fun to hang with.
Here’s something that’s not normally newsworthy, but this year I thought it was. The t-shirt they gave out for SWW09 was far better than usual, in fact I liked it so much I wore it to the Magic Kingdom one day. Along with my SW hat. What a freaking SW billboard. Kim is obviously the one taking the pictures here.
Anyway, the shirt actually has little pieces of cloth on it, giving it texture and a little depth. It didn’t have any food on it at this point.
Matt,
although we weren’t in agreement on the SW interface that I think it’s still the best and the direction it’s going I like, I think you’re doing an excellent job in stimulating the development team of SolidWorks to follow the interests of users .
We hope in future the use of Direct X, the unification of the RealView, PV360, PW a better direct editing (not the ST in NX, which is useless) and more tools for manipulating surfaces.
Just as a quick comment, great surfacing starts with great curves. This is one area that Solidworks just doesn’t seem to do all that well. Along these lines, which I have already expressed frustration with, referencing existing curves/points/sketches is a key. Points are almost useless in Solidworks, but a very key in some of the more robust surface models I have built in another software package. 3D curves are great when they work, but they just don’t seem to be all that robust to model changes. My work requires both complex surfacing and multiple sizes.
The other part of surfacing that seems to need some work is the robustness to change. Often times Surface Trim features fall to pieces with the smallest of model changes. References are either lost or reversed, neither of which is helpful. I like to be able to show people how robust a surface model can be, but SW has not allowed me to do that yet.
I do have to give props to the Filled Surface though. This is a great tool that simplifies things quite a bit when you can use it.
That should be a good start!
Eric
Lofts make me crazy. It is an annoyance that a loft with 3 or 4 sections will initially have one or more terrible twists. I alwyas use guide curves and that tends to straighten things out. It is difficult to create nice visually flowing shapes for aircraft nacelles and fairings. Most of these shapes are not tightly constained by the aerodynamics. I want the motion of a highlight to be smooth. I need more control of lofts. I want the loft to exactly match the sections. I want the loft to have a shape that also touches the guides, but can usually accept some small deviation relative to the guide curve. When two sections are convex, I expect all sections between these to be convex, and have an intermediate shape. The shapes that I use would be well served by the intermediate shapes being a linear combination of the section curves, and stretched independently in the two dimensions. Guide curves should not make a crease or concave section unless the sections have a crease or concavity. When the sections and guides are symmetrical the loft should be symmetrical. Lofts intersecting lofts are common with airplane shapes. Solidworks often fails to merge bodies if two lofted shapes share a guide curve. I would be happy to share the model of a conceptual flying boat that has several intersecting lofts.
Regarding your discussion with the VP of Customer Service:
“It is clear to me that none of the suggestions we had are even going to be considered.”
That’s 100% not cool. I’ll try v2010 before blindly buying this time.
I really like using sketched curves projected onto sketched curves to obtain very controllable guide curves. The airplane fuselage is largely controlld by an inboard profile, and a plan view. I like to make the big changes at this top level sketch, and have all of the shapes follow. This little seaplane has a split personality, the bottom is a boat and the top an airplane.
I need a ftp site or email address to send the seaplane model. I am open to all comments regarding the model and the design.
Rick,
I’d be very interested in seeing the model. Plus, I’d recommend using Boundary instead of Loft, especially where accuracy to existing curves is important. One of the things I learned this go round at SWWorld is that the curve features tend to have wider tolerances than edges, and boundary is more accurate than loft. Thanks for the input. I agree with your expectations from lofted features.