Sorry, but Sun offends me only just slightly less than Microsoft. One only has to witness how they are attempting to insert Java throughout all sorts of media and network standards (thereby ensuring all implementations pay the "Java tax") to see that Java's "openness" is highly overrated. McNealy'd like nothing better than to turn Sun into another MS (bad behavior and all), and certainly hasn't skimped in applying monetary and political pressure to get his chance. Anyone who believes Sun (or Oracle for that matter) gives a rat's ass about the "public" in their fight against MS is delusional.
The idea of designing and planning the IL code around efficient JIT operation, and giving it less language-dependency is a great modification of Java's approach. It shows that Microsoft has put quite a bit of thought into how to swat Sun aside, by addressing the problems people want fixed, as opposed to endless oft-incomplete "augmentation".
The runtime libraries are also very well done. They seem to have a far cleaner notion of heterogeneous usability than AWT/Swing/etc., and provide lots of useful stuff without much bloat. Again Microsoft is showing some real talent in how they've optimized the design and selected what made it into the runtimes. Compared to Sun's extremely politicized and far-from-open Java evolutionary mgmt., MS has again (and as usual) demonstrated their superior ability to "iterate towards optimum." Perhaps if Sun had actually been open about Java's development, this situation could have been avoided.
Overall, MS has shown a much greater "clarity of implementation" in.NET than Sun ever did in Java (instead preferring to use Java as the kitchen sink). Sun deserves to get spanked for how they muddled and confused their intents with Java. We won't get into Sun's primary apparent use of Java as a "licensing money tree" which they come back to far too often.
Oh the irony...doctor, heal thyself first. Did you read the books? I ask only because you grotesquely misinterpreted just about the entire content of "Dune Messiah" and "Children of Dune". It would be easier to believe you didn't actually read them, than to believe you got that from reading them.
There's my 2 cent version of a book review.
You're charging too much, and I'd hate to see the "expensive" version...:P
Look at the options presented for the product. Nowhere are support issues available for discussion, etc. that I see.
I'd strongly recommend that anyone buying into such gadgets at least make a cursory check to ensure that the company is prepared to support existing customers...and more importantly, how those existing customers feel about said product.
Yeesh.
Why shouldn't Transmeta be held to the same standards, though? Creating "special" benchmarks is precisely what Transmeta wants, I suspect, but that doesn't mean it's the right thing to do.
Is it reasonable, by extension, that every processor manufacturer who comes up with some nifty power-saving feature be allowed to change the playing field to suit themselves? I certainly don't see the consumer coming out ahead in that scenario. Tuning the benchmark for Transmeta completely misses the point. Even though the tests aren't repeated, the same segments of code are used over and over again...and the code in question is designed to be the typical mishmash of user needs. If Transmeta cannot adequately compete on that field, then it's something they need to fix.
If you look at the review of the similarly-priced ($100-200 more) Sony ultra offering using a 600MHz Mobile P3, it definitely draws into question what the value of Crusoe is in context. The P3-powered ultra weighs slightly more, but totally runs rings around Crusoe in the benchmarks, and doesn't appear to suffer notable battery hardship in doing so.
In other words, what precisely is the value of Crusoe if it cannot compete on performance, and doesn't represent a compelling win on battery life? I'm starting to wonder if Transmeta should have taken a bit more time...
Slashdot Mirror
The idea of designing and planning the IL code around efficient JIT operation, and giving it less language-dependency is a great modification of Java's approach. It shows that Microsoft has put quite a bit of thought into how to swat Sun aside, by addressing the problems people want fixed, as opposed to endless oft-incomplete "augmentation".
The runtime libraries are also very well done. They seem to have a far cleaner notion of heterogeneous usability than AWT/Swing/etc., and provide lots of useful stuff without much bloat. Again Microsoft is showing some real talent in how they've optimized the design and selected what made it into the runtimes. Compared to Sun's extremely politicized and far-from-open Java evolutionary mgmt., MS has again (and as usual) demonstrated their superior ability to "iterate towards optimum." Perhaps if Sun had actually been open about Java's development, this situation could have been avoided.
Overall, MS has shown a much greater "clarity of implementation" in .NET than Sun ever did in Java (instead preferring to use Java as the kitchen sink). Sun deserves to get spanked for how they muddled and confused their intents with Java. We won't get into Sun's primary apparent use of Java as a "licensing money tree" which they come back to far too often.
Hehe, that's like saying Alia was a little conflicted about her loyalties.
This just about made my day. Thanks!
Oh the irony...doctor, heal thyself first. Did you read the books? I ask only because you grotesquely misinterpreted just about the entire content of "Dune Messiah" and "Children of Dune". It would be easier to believe you didn't actually read them, than to believe you got that from reading them.
There's my 2 cent version of a book review.
You're charging too much, and I'd hate to see the "expensive" version... :P
Look at the options presented for the product. Nowhere are support issues available for discussion, etc. that I see. I'd strongly recommend that anyone buying into such gadgets at least make a cursory check to ensure that the company is prepared to support existing customers...and more importantly, how those existing customers feel about said product. Yeesh.
Why shouldn't Transmeta be held to the same standards, though? Creating "special" benchmarks is precisely what Transmeta wants, I suspect, but that doesn't mean it's the right thing to do. Is it reasonable, by extension, that every processor manufacturer who comes up with some nifty power-saving feature be allowed to change the playing field to suit themselves? I certainly don't see the consumer coming out ahead in that scenario. Tuning the benchmark for Transmeta completely misses the point. Even though the tests aren't repeated, the same segments of code are used over and over again...and the code in question is designed to be the typical mishmash of user needs. If Transmeta cannot adequately compete on that field, then it's something they need to fix. If you look at the review of the similarly-priced ($100-200 more) Sony ultra offering using a 600MHz Mobile P3, it definitely draws into question what the value of Crusoe is in context. The P3-powered ultra weighs slightly more, but totally runs rings around Crusoe in the benchmarks, and doesn't appear to suffer notable battery hardship in doing so. In other words, what precisely is the value of Crusoe if it cannot compete on performance, and doesn't represent a compelling win on battery life? I'm starting to wonder if Transmeta should have taken a bit more time...