I am not a beta tester but am aware of the NDA all must sign. Therefore it's possible that nobody who knows will be allowed to answer my question. However, I am wondering how they can truly beta test a software that is supposed to run on an OS that the software doesn't run on? And also test it on an OS that hardly anybody has loaded on their machines?? (not to mention driver issues)
Seems obvious to me that the only beta testing of 2008 that is being done must be on XP. Therefore, how can anybody even be talking about SW on Vista is there is no way to prove that it will work in a production environment??? This would seem to indicate that the first true version of SW for Vista will have to wait until at least a later service pack of 2008, if not SW 2009.
As a result of the uncertainty over lost productivity, I'll bet SolidWorks users stay on 2006 and 2007 a lot longer this time before upgrading to 2008.
...also you might also look at the token prereleases and consider beta testing was a farce again...I wouldnt bother about violating NDA either...after what SW themselves chose to strategically leak thru the blog squad and the draft 'whats new' its been rendered fairly pointless for months....
Ultimately it is up to SW to test their software and provide it to users in a finished and usable form. Over the years we have come to expect to have to test it ourselves. It is rather surprising to me that the OS makes that much difference in how SW operates. But it does.
Having been around for a while (like you) I went over in my mind the history of Beta testing (help me here with dates.)
When SW was marketed circa 1995 the company was very tight lipped about any new feature or functionality. When released it was solid ( I rarely found problems). Problems were handled in Concord and that is were you called for tech support. Pro/E seemed to be the paradigm for what the software was supposed to do and at least in my mind Pro/E was the standard to which SW should be evaluated. 166Mhz was a screaming fast processor at the time. From 95 to 98 there were few changes to the user interface and some basic functionality that was still being sorted out like having a single mate tree in an assembly.
By 1998 SW was starting to have a few show stopper problems and for me performance was starting to be an issue (300 Mhz was about the fastest processor and it wasn't adequate for mold work). I started asking my VAR when he thought the next SP was ready to use. SW would never release vaporware, bucking the general trend in the software inducstry.
By 2001 SW was starting to involve power users on a limited basis in testing and product definition. The 3 Amigos met Biasotti in the lobby of SW. Beta was still not open to the public. 1 gigahertz processors were starting to be available and were still not fast enough. SW was using statistics to determine what to fix and whether the product was acceptable in the marketplace. There were small inklings of future directions at SWW.
Around 2003/2004 SW started to open up beta testing. At first it seemed to come out of the trend of having to ask the VAR whether SW was really ready for production. SW was letting users identify problems and leaving problems that user's didn't identify as being show stoppers till last or never. 2003 was a major shift in functionality in SW with the addition of multi-body support. This had been in UG and Pro/E for years and was the last really great improvement year in SW. SWW was being used to "vote" on new functionality. Feature bloat in response to Inventor and SolidEdge seem to become a driver in new functionality.
Currently it seems that SW has started to depend on schedule and beta testing to wring out problems and release the product on time. SW regularly releases vaporware at SWW now. Question: "At which SWW did SW state that they would add some limited backward compatibility to the software?" Pro/E is no longer the paradigm, Inventor and SolidEdge are the paradigms.
What is wrong with such extensive beta testing? I would say skewed statistics. People who are actually using the software probably don't have the time to spend on beta. I know, I have spent hours at it. It is hard to get work done and beta test and do a reasonable job at both. Most beta testers will not have the latest cutting edge hardware whereas SW always seems to. Vista is a good example.
Well you have been a user longer than I have but I tend to have a slightly different perspective on this. I am afraid I am somewhat more cynical than that. To me beta testing is pretty much baloney - not because testers haven't tried in their sincerity but because SW don't do anything meaningful with the info prior to release. The number of bugs in a release has gone up not down. Honestly, how can they release PR1 and then PR2 a week later and have people believe in it. Same as last release when they skipped through it to meet their feel good launch schedule.. I'll bet many people didn't even download PR1 let alone test it before it was gone. Are we really supposed to believe they fixed things in-between? We probably would have more success finding weapons of mass destruction. To me beta testing has been both a free lunch on the good will of users when they should have paid a team to do the job and a clever but empty PR gesture. Lets face it by the time a release gets to testing its pretty much set in stone and the bulk of fixes are going to arrive about sp3 regardless. Whatever Matt might think he achieves in complaining about the interface for example its pretty much pointless and he should know that. I well remember the complaints in the past about playing the 'we got it too' game with SE and Inventor and half finished half baked features and its all the same deal today. There are still major problems with SW management grasping vital clues about being in business to serve customers rather than farming customers and keeping pet blog squads. We are talking about an engineering tool here not having something in the showroom to dazzle 'em with every 9 months.... ..but all old comments... and why I stopped buying.. ;o)
I looked hard at both prior comments and, thought of my experiences since 2000-2001 with SolidWorks and with other "tech" on the bug side.
Microsoft has been so susceptible, I had to make major work change to keep SolidWorks running happy as much as possible. That means limiting XP Pro to ONLY Swks and MSOffice and Nothing Else, including the Internet.
I have had to avoid all SolidWorks early releases until the software gets to about SP4 before I load it, and things then don't seem to have major problems or anything but small #s of crashes.
Then when I compare Microsoft & SolidWorks with another software intensive platform I use, I wonder why MS & Swks can't do the same.
Apple's OS's have been rock solid for 5 years for me. Once in awhile an application quits and then restarts w/o problem with Apple. I don't wring out every last feature, but I don't recall a show stopper, and have had maybe 1-2 OS crashes in 5 years, that caused a reboot...that's it.
Then Apple announces the iPhone, and the industry leaders like MS & Palm's CEOs trash Apple for not knowing what they are doing, and the pundits write what appears to be a hundred articles about why it will not be right, acceptable, functional, bug free, safe, for so many dreamt up reasons.
The reality of the iPhone was? I have used it for 2 months and it is the only phone I never read a manual. It is the only phone where every feature I tried worked as it should (though I didn't try every option). Crashes? I crashed my Trek and the iPhone kept on working. That crash hurt me. The iPhone is still solid.
It is possible to design good code and release it when you say you will, and the customers will pay a premium to get good systems, with good code. Apple has shown it, and their growth rate of triple the industry shows they are making hay off of their skills.
As this decade progresses, I predict that MS is going to slowly get edged out, as it continues to believe an Operating System is going to continue to be premium priced and that the OS must "look cute", and that an OS must be humonguous and please everybody's wishes with every release and be DRM'd to death for every copyright holder, while at the same time MS elbows the standards orgs to adopt their rediculuously complex "OOXML" or whatever other standards, to try to lock in the world to MS products.
If MS were smart, I would say they take the XBox approach and start designing high end PC boxes that people scramble to buy to match Apple's path that has been shown to work, but that is just me, a stupid user speaking. Mind you, MS would not want to copy XBox quality in PC boxes or they would go BK.
Then SolidWorks.
User interfaces must be more thoroughly planned out to be the best and most intuitive. Reading 1000 page manuals with every other major release is the pits on productivity.
Major Release once every two years
Major Release must be free of major bugs
Major Releases must have functionality pulled if not solid.
No Service Pack goes out without extensive developer testing.
Manuals on DVD need to be provided
The Most modern code writing, organization and debugging systems must be employed.
The best people must be hired in the "write" places.
Toyota and Apple did not get where they are by taking shortcuts. (Interesting analogy: Microsoft calls file aliases "shortcuts", and that seems to ring true with me about Microsoft).
Companies that don't emulate the execution shown at the top of their industry are going to ultimately face users who abandon their products.
I hope SWks gets its act together. I'm sticking with Swks 2006 for now, though I may hit 2007 soon. We'll see how a trial run goes first.
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.