Even mid-size or smallish companies with 6-12 people can get into the inefficiency groove, and just say to themselves "Why fight the CAD software issues and get a black mark on my name in the personnel department?"
I know one company near me in my line of medical work, that kept using SolidWorks 97 or 98 clear up to 2004 or 2005 last time I heard. That meant they were living with a lot of things that later versions cleared up, though indeed new glitches arose in later editions.
My belief is that if SolidWorks had more bug free major releases, that the company I referred to would have probably updated every year, and everyone, the customers & SolidWorks would benefit.
If I only worked with internal files, I would still be using the late sp's of sw2000. Considering it ran on a PII whatever machine, it was incredibly fast compared to recent versions with less crashes and freezes. It seems like ever since then, bugs and problems have steadily increased and speed has decreased with every release and every sp.
It is all well and good that SWX might use such a tool as Alloy to find and fix the system design. However, SWX is made up of numerous modules like parasolid over which they have no immediate control. And then there is Windows which they not only have little control over, but is shrouded in proprietary secrecy. Add to that SWX being a little fish in MSof'ts big pond and it becomes apparent that SWX would have a long row to hoe to use such a tool. And Alloy won't find errors in the specification for what SWX does. It would not fix a problem with, say, inconsistent icons on the toolbar (something SWX recently fixed).
On the other hand, if SWX had their own code and ran it on their own OS (Linux*) then such a tool would make emminent sense. This was one of the big arguments for ProE back when SWX was small potatoes.
You might ask why I state that Linux is SWX' code. Simple, since it is open source when SWX finds a problem in the OS they can fix it, place it in the public domain and it will become part of Linux. Anybody can do this. Little fish becomes shark. Compare this to the current scenario where SWX has to beat on MSoft to get them to acknowledge a problem. And there have been problems, as those of us who used SWX on NT well know.
If Swx was to bundle their own version of Linux with their software, that would be great. I don't think the average swx user would miss Windows too much. I believe you're right about whether enough of the modules would be available for the new OS. I would imagine that would be a bigger challenge than writing the base program. Parasolid was probably developed on some flavor of unix, AIX, whatever, but it may no longer be produced that way, and many of the more modern mods were probably developed Windows only.
I think if SW is going to go in any direction with any other OS, it would be OSX. Linux install base is servers and hobbyists. Most users and IT groups aren't going to go for Linux as a desktop OS. I don't think the prospects have changed much for Linux over the past several years, you still don't see casual computer users running it.
I'm not sure it unreasonable to suppose that if a OS X port is made, that Linux may not be far behind or even in front. OS X is based on BSD Unix, and so it very closely related to Linux. If SW were ported to Linux using the right tools, OS X support might be had almost for free, so to speak. Also, many graphics workstations run Linux, so I don't think the installed base is quite so limited as you do (though it is severely limited compared to Windows).
But there are many obstacles, either option is highly unlikely unless Vista is a major flop. Personally, I think the system requirements are far too high for what Vista will actually deliver.
I don't think most people working professionally with software like swx, cam packages, photoshop, etc would be considered casual users. If they could get a robust setup, they would dump Windows, etc in a NY second.
Spend an afternoon with a reseller tech support or training group, and I think you'd change your mind. By casual, I mean someone who doesn't install OS, hardware that requires a screwdriver, manage the registry or memory settings, etc.
There are already rumblings in corporate America about the pain of switching to Vista which include NEW security vulnerabilities, which many companies simply will not tolerate for the first year or two until the bugs of Vista are mostly squashed and they are prepared to deal with end user installs & support.
Sooner or later a non-proprietary OS is going to take over. SolidWorks is going to have to follow the crowd, though, as all their partners are going to have to move over to that OS (whether it is Unix, Linux, or Mac's OSX version of BSD Unix). Lots of their partners probably already have solutions which go on Unix systems, so some partners are already "there".
Corporate America is already happy and satisfied that Unix runs solid and is relatively secure, and that is a major bonus, along with the fact that there are hundreds of thousands of programmers and SysAdmins using Unix today, who understand Unix, and deal with it every day.
Gut feel says Unix will win out, if SolidWorks were to switch.
All I want is an end to the BS which WinXP Pro has given me, and the stuff I read about with Vista Beta notes.
When we SolidWorks users pay anywhere from about $5k-10k to start with SWks software plus yearly 4 figure maintenance, pays more for other add-ons, buys a high end PC for another $4k and has to have all his other thousands of dollars of software, we don't want to have a bloody $400 OS with buggy problems to deal with.
Bad $400 OS software, cripes. And then it makes the other $10,000 plus pile of hardware & software a real trick to run. It is a plain crock to have a crummy OS that slows us down.
MS has let me down. I won't speak for anyone else.
I can get XP Pro to run SolidWorks just fine, but I can't keep it "clean" if I allow it to run on the Internet, so I don't. That is a major headache.
I don't argue with your distaste for big lazy corporations who would rather win competition through sheer size rather than by producing a superior product, but I disagree with your analysis of what's going to happen.
Don't look to Corporate America as your OS savior. Corporate America is not going to be the one who makes the switch to Linux. Linux has been available a long time now, corporate IT departments have long had the opportunity to switch, and are not doing it in droves. People have been saying Linux was going to take over for years, and it just doesn't happen. Why would someone with as much money as Bill Gates allow open source software to eat into his profits? He would buy it himself if it got big enough to become a threat.
Corporate America is what made Microsoft so huge. The only thing that will challenge Microsoft is another mega-corporation. Linux cannot get large specifically because its camp is so divided and decentralized.
I understand the individualist idealist, the Robin Hood mentality, the bandwagon that routinely bashes whoever is on top, and the genuine desire for better software, but as long as human beings act like sheep that are herded together by large corporations, we will only have privately owned corner hardware stores (or operating systems, or gas stations, or crafts shops, or hamburger stands, or music, or regional identity) until it becomes profitable, when it will be taken over by a capitalist who thinks big.
I'm not an advocate of the corporate bulldozer, I just recognize it as a fact of life in this century in this country, and unless you live in a hollow tree in north western Montana, you've helped buy into it.
Absent crap and corruption of the OS in any one of the dozen common ways, starting with Registry, adware, spyware, IE6 issues, etc, is what I mean by "clean".
I run behind a firewall, use Bit Defender Pro, & such when I go online, but I generally do NOT allow it online unless a particular upgrade or patch can't be downloaded elsewhere on my Mac. Doing this has kept my Dells totally free of downtime, and that is terrific. But a real world OS should be able to go online routinely and not have such problems, yet it doesn't seem to work out that way with XP Pro.
I have to help others occassionally fix their XP Pro issues, and geesh. Who has the time to burn to do these things?
We are in the new millenium, and OS's ought to be there, too. Unix & Linux seem to be delivering on being trouble free. I only use Mac OSX's form of Unix, and to date it has been rock solid, even though it is online a lot.
I just installed Windows Server 2003 at work on two machines, one a database server and the other a domain server. As with Windows XP, these systems do not support the latest hardware at install time without major screw driver work. Linux does. We had a Windows certified tech come in to install SQL Server. Seven billable hours later we had it up and running and a rudimentary domain also. It will take weeks to learn the intricacies of running a domain server the way we want (because of having to plow through Windows obscurity). The tech was of course dyed in the wool Windows at the beginning of the day. I took him on a quick tour of the Linux server that had been holding up our network for a year. He began to realize that a configuration file like this:
$TTL 1D @ IN SOA engtran.com. root.engtran.com. ( 2003050207 8H 2H 4W 1D ) NS xxx.xxx.xxx.xxx NS xxx.xxx.xxx.xxx TXT "bogus engtran for intranet"
localhost A 127.0.0.1
while somewhat cryptic at first glance fully documents the setup of a dns zone unlike the series of mouse clicks and the arcane presentation found in W2K3 DNS server. The fact that this setup, once working can be copied and moved to an entirely new Linux box on a memory stick is far superior to have to re-click everything on a new install. So from a maintenance and setup standpoint for IS, Linux has many advantages. If you don't believe it look at the second line in the zone file. It was written in 2003, 3 years ago and has survived an upgrade or two.
Add to that that most people have not seen the GUIs available or the flexibility thereof and the relative obscurity of the causual user base is explainable.
But I don't hold much hope for SWX going to Linux unless they went GPL and that is never going to happen.
Ok, that's what I thought. Here's my experience so far. This machine I'm working on serves all functions, SW, email, print, storage, backup, all of it. It's been runnning for 15 months now with no need for a rebuild other than correcting a misjudgment in partitioning choices (I would have thought
100 GB was enough fora c:). Trend Micro has only found a few items to complain about, all cought in Outlook 2003's junk mail folder and quarantined there. I'm trying to say that running XP is better than it was two years ago, for a user who is careful what gets clicked on.
I hope I don't sound like I'm advocating XP as a good OS, I'm just saying that it's not impossible to keep it running fairly well (for XP). I don't think it's necessary to keep a separate machine just for Internet access.
Individuals in small shops just have to pick what they can with the limited hours they have to try to get the job done efficiently, or they run out of time and don't deliver the goods, and run out of money and close their shop.
Efficiency is the name of the game.
My pick of an efficient way is no doubt different than others.
Some people are using only Windows, some 2 Windows machines, some Linux and Windows, and I use Mac OSX and Windows.
If I were to ONLY use Windows, I would still have 2 XP Pro machines to maximize my efficiency.
When I get my hands on a dual core system, I'd like to experiment with Linux in a virtual machine. I think it would be prudent to start learning Linux, and I've been curious about it for years. It would also provide a handy sandbox for less reputable software.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.