This information is from memory, from a few articles that ran in magazines about a year ago -- I want to say either Forbes or Fortune, but I'm not sure, as well as some of my friends' car stereo magazines. Anyway, at the time XM was still pretty new, and Sirius was just beginning to take subscribers.
First of all, XM is a Sony operation through-and-through. The main reason XM is doing so well is that Sony has pumped a bunch of money into it, knowing up front it'll run at a loss for a long time. On the other hand, Sirius is an independent business with all the attendant risks and concerns. Unfortunately Sony is a giant, monopolistic bastard of a company who pimps cheap junk at ridiculous prices, so you can guess which side I root for in this battle.:)
Second, there is the satellite problem. Sirius placed a small number of satellites in very high orbit. This means each satellite can cover a relatively wide portion of the Earth's surface. Also, these satellites are in a more stable orbit so they have a longer life expectancy. Launch costs were higher, however. Conversely, Sony opted for low-orbit satellites. This means the launches were cheaper, but the life expectancy is lower, and each satellite has a much smaller area of coverage, meaning Sony had to use significantly larger numbers of satellites.
Worse yet, the XM satellites are unexpectedly having problems holding their orbit, so they're using fuel at roughly 3X the expected rate. This is one of XM's largest problems. Sony faces having to replace the satellites YEARS earlier than expected. I don't remember the exact figures to replace the satellites, but the price tag was vast -- like $700M or $1B or something equally insane.
Third, there is the ground station plan. Both systems use a network of repeater stations. These (combined with receiver onboard buffering) ensure you have a steady signal when you drive through an underpass, for example. Amazingly, Sony only has 75 repeater stations in the US, all near major cities. The Sirius plan calls for 2,500 repeaters spread somewhat evenly throughout the country. I don't recall how many Sirius currently has, but it's already much more than XM.
Fourth, of course, is the question of content. Sirius costs a bit more than XM, but it's also largely commercial-free. Sony charges you a not insigificant amount for XM, and then they sell advertising on almost all their stations anyway.
Fifth is the availability issue. Since XM is a Sony-funded effort, it was available everywhere almost from Day One. However, Sirius has a ramp-up plan which targets major cities first, then rolls out to everyone else over a 12-18 month period. This may be important to frequent travelers.
Sixth and finally, speaking of frequent travelers, another post pointed out that people like truckers are a prime target market for satellite radio. Interestingly, there is another important target market mentioned in the articles, but overlooked here so far (I think) -- ships. Merchant ships and cruise ships are potentially a HUGE market for satellite radio. It turns out that XM was very heavily focused on land-based use, and their offshore coverage is largely accidental. Because of this (and XM's use of low-orbit small-footprint satellites) XM coverage doesn't extend very far into the ocean. However, Sirius took that market into consideration when they planned their service, and their coverage extends quite far out to sea. Again, unfortunately, I forget the exact numbers, but the difference was serious, like 20 miles compared to 400 miles.
It's my opinion that Sirius is the better technology for these reasons and others (even ignoring my intense dislike for Sony), but we all know how often Joe Sixpack pays attention to little details like that.
Again, that was all from memory based on information I read almost a year ago, so please don't beat me up if I got any of it completely wrong (and please correct me). Hope you found it interesting.
Good old networking under Windows 3.1...
One day this error message appeared on my screen:
ERROR: NETBIOS has become Twizzled.
I clicked Ok, and everything continued working normally. I searched the hard drive and couldn't find that text anywhere on the machine. I contacted Artisoft (we were one of their larger customers) and they insisted it wasn't from LANtastic. So I contacted Microsoft (back when they had support people you could talk to) and they also denied complicity.
To this day I'm not exactly sure what "twizzled" means. Never saw the error again.
I find it amusing that the highly-UI-aware Apple manages to produce some of the most bizarre non-standard Windows applications I've ever seen. Talk about "custom controls"...
Yes, it does invalidate the point you made. Your point was that "Einstein showed something". But he didn't. One can show (=prove) that pi must be irrational, but one can't show that objects can't accelerate faster than light (even if it seems plausible and likely). The distinction is particularly important for physicists to keep in mind; too bad that it is lost on many of them.
I see I have to get nit-picky too.
My point definitely was not that "Einstein showed something".
The point I made was based on accepted theory which SUGGESTS (happy?) something may travel at speeds in excess of the speed of light, and that only acceleration to and/or beyond the speed of light is impossible because the energy required would be infinite.
The original article contains an assumption on the author's part that the theory requires that the speed of light is an inviolable upper boundary. My point was simply that this is an inaccurate interpretation.
Therefore, my point is not invalid. The worst you can say is that my choice of a single word was unfortunate. As another responder noted, perhaps this subject requires nit-picking to a degree which makes such distinctions important, but somehow I suspect my/. response isn't going to influence the researchers involved.
So, as I noted previously, I concede that Einstein didn't SHOW anything in the sense that SHOW can be interpreted as providing incontrovertible evidence (and I suppose I could argue that point, too, if I didn't have a life to get back to). Nonetheless, accepted theory does allow for travel beyond the speed of light, and therefore the original article text's assumption that the speed of light was an absolute upper limit is based on an incorrect interpretation of the applicable theory.
Hopefully you are happier with this over-anal-yzed wording.
You're correct, which doesn't invalidate the point I made, does it? That makes the value of your contribution... well, let's just say I wish the moderation system included "whiny anal nitpicky bastard" as an option.
All Einstein showed was that nothing could accelerate beyond the speed of light. Nothing prevents something from moving beyond the speed of light provided it always moves faster than light.
What's truly astounding is that the same managers, who were at pains to hire the brightest people they could find, think that those same people won't figure out what a fraud that is.
No, what's truly amazing is that many of those managers don't realize what a fraud that is. After many years in Corporate AMerica, I've come to the stunning conclusion that those kinds of windbags generally think they're actually doing a good job.
Everybody is forgetting the "accessory" I find I use the most -- a big old ziplock baggie full of spare screws, motherboard mounting posts (and those square snap-in mounts), washers (paper and metal), and just about every other computer-related fastner you can think of.
How many times have you cracked open a case to discover that every drive was held in by a single screw? Or that screw that the owner has stripped and is relying on it being wedged in just right to hold the $400 video card in place?
Of course, it's easiest to acquire these items by simply collecting them over time. That would be a cool thing for an online retailer to sell. "500 piece miscellaneous computer fastner pack... $7...":)
Oh yeah, in case nobody said it: zip-ties and velcro tie-wraps...
Since VB does not run on Unix servers, Linux servers or Mac OS X servers, I can only presume that your work experience is totally Microsoft. Certainly your "limited" view of why Java is an effective programing environment for server-based software leads me to believe that.
Yes, I worded that part very poorly. My point was that it has many of the same attractions that VB has -- easy to learn, easy to deploy, etc. No wild pointers bringing down the service... one of those languages where anybody can start cranking out code without knowing much about how stuff actually works. (I'm not saying that's a good thing.)
I agree that HTML has killed VB on the client, too, but that's not really relevant to the discussion. Indeed, there's probably a larger market for COBOL guys now than VB guys (a guess, but my company sure has a shortage).
Java has never delivered on it's client-side promises, and it never MADE any server-side promises, it's just that the server gives the programmers enough control that the "write once / run anywhere" myth doesn't bite them in the ass and nobody cares about how difficult it is to build a decent UI in Java.
Java on the client was mainly killed by plain old HTML and round-trip processing (CGI, ASP, forms, etc), which proved far superior for quickly and easily building decent, usable UIs, and which downloaded to and rendered on end-user machines much more quickly.
The really stupid thing is that Microsoft was probably Sun's best chance for keeping Java alive on the client. MS had the fastest JVM around for a LONG time, and they had the most bug-free JVM for a long time. (I wrote Java apps for a number of years and had to constantly test on about a dozen JVM/JITs, I even recall the first Sun Java event where Sun people were praising and recommending the Microsoft JVM.)
Microsoft probably could have killed Java, and it's reasonable to assume they thought about it -- although I personally disagree that J++ was an actual attempt to do so (it was a kick-ass environment, and I produced vast amounts of full-compatible code with it) -- but the simple fact is they didn't have to, because Java never really got going on the client. Java lives on in the server-side world as a kind of VB-for-people-who-hate-Microsoft, and that's about it, but MS didn't kill it.
For those of you trying to figure out what "negative refraction" actually implies, the article at the URL below has a pretty easy-to-understand explanation of the key characteristics.
The use of apostrophes is an obvious improvement over the quotes used by BASIC:
PRINT "Hello, Python."
Now THAT's what I call unacceptable syntax.
(Of course I know the difference, it just struck me as amusing.)
Just a couple of points in the interest of basic accuracy:
You also neglect this technical problem in XP: "If you say no to some of the requests, some functions of Windows XP will not work (such as networking)." If you deny internet access to many components, XP will cease to function properly. Did you notice the long list he had of components that needed 'net access? Windows Media Player!?! That's a technical flaw that's borderline malicious.
Untrue. I agree that a stupidly huge number of apps and processes attempt to access the 'net, but it isn't true that XP will cease to function properly if you deny that access. My wife's external firewall is configured to automatically deny access to everything except to a small handful of specific apps/ports/addresses, and XP runs just fine. Yes, including Media Player.
There is also no technical solution things like "Run DLL as an app" not telling you which DLL needs to be run. These programs should not be calling home unless they need to in order to function properly for the user's benefit. They way they work now is simply frivolous.
I do agree with you here in principle, although it isn't technically impossible to determine what DLL is making the call, just difficult. Many of the tools from SysInternals will show you all the nitty-gritty details of what DLLs are in use by each process, for example. Interestingly, Tiny Software's Personal Firewall is still able to block multiple apps using RUNDLL or the generic "service" process by maintaining hashes as identifiers. Unfortunately it's still up to the user to track down what specifically is trying to make the connection, though.
Also it should be noted that XP really doesn't "call home" as you said. At least, not in the sense that you wish to imply. If you pay attention to where the calls go, they do perform some useful task in almost all cases. (The value of these tasks is a completely separate question.) The real problem is that the user isn't given any choice in the matter, or even told that it's happening. But none of the connections I've seen or that I'm aware of were ever specifically a "call home" facility. Yes they might be tracked that way, but they aren't raw "call home" connections as you imply.
The parent should be modded up. Wish I had points today.
If the original poster isn't just lying or wasn't trippin' balls, I'd guess he saw a B2. I've seen B2's pass overhead at night, and they're exactly like what the article describes -- enormous black triangles that make no noise and appear to be moving very slowly (since they're frigging enormous).
I've also seen B2's pass from the side at airshows, and like all very large aircraft, they appear to be moving impossibly slowly. It's an optical illusion, of course, as they have stall speeds well over 100 MPH (a B2 is probably around 115 MPH). Having grown up on military bases I was very accustomed to staring at 100-ton C5 Galaxy transports coming in for landings at what appeared to be liesurely single-digit speeds.
This whole giant secret blimp thing sounds pretty silly, though.
First of all, Microsoft hasn't stopped supporting VB. VB5 support will continue through 2004. VB6 support will continue through 2008. And of course, VB7 is VB.NET, which was just released in February. Although the number of VC++ programmers surpassed the number of VB programmers a little over a year ago, VB is still a huge product for Microsoft. To suggest that they would stop supporting it is ludicrous.
Besides, I didn't say VB support wasn't important, I said it wasn't the definitive litmus test everybody says it is for early adoption of.NET. Remember, the main topic is why.NET is being adopted more slowly than MS expected. If all the VBers in the world made the switch overnight, the sheer size of the Framework (over 100MB) would prevent them from forcing a rapid wide-scale deployment. Many VBers have difficulty making the case for the much, much smaller runtimes in previous versions (what's VB6? something like 6MB now?).
.NET has been marketed as being best suited for writing web-based applications, but the point of this thread is that any such statement is misleading. Web-based apps are simply one relatively small facet of what.NET is capable of doing, and if you'll take a moment to look through the Framework SDK documentation, you'll find the vast majority of it is centered on writing regular stand-alone Windows programs, building controls, database I/O, component-oriented services, and other things which aren't even remotely web-centric.
.NET can do web-based apps well, but it's instantly obvious to anybody familiar with the Framework that not only are web-centric apps not the central focus of.NET, but that they're really just a bullet-point in the list of its capabilities.
Heck, the machine I'm using now has a whole slew of.NET-based services (in the NT Service sense of the word) running, I've already written two fairly significant.NET applications which aren't web based at all, and I'm in the middle of a third which has the ability to communicate via TCP/IP but definitely isn't web-centric.
Sorry man, but I think you really missed the boat on that one.
I think somewhere in there you missed my point. I don't think all the slogans in the world could sell something like.NET because it's simply too large and complex to lend itself well to a simple but accurate description. Additionally I disagree that cool bits were marked alongside uncool bits -- I believe the uncool bits were the ONLY part that received marketing attention, and that is the main reason for the failure.
While I agree the VB compatability question does play a role, I don't think it's as important as you suggest. With each rev of VB, the previous version (sometimes the previous two versions) remained in widespread use for quite a long time. This has become normal in the VB community.
I believe the actual problem is that Microsoft and particularly Microsoft marketing has done a really poor job of explaining to developers what.NET really is and what it has to offer. Most developers simply don't know what.NET really is all about. I've been to MANY.NET-related shows, seminars, conferences, and other things over the past couple of years and even near the end of the show, people are wandering around confused -- often because what they're hearing has NOTHING (or very little) to do with what the marketing droids have told them to think about.
The complexity is really a side-effect problem. If I thought.NET was only web-services-related, which would be an easy conclusion to draw based on the marketing, and I was then faced with the thousands of classes which comprise.NET, I'd start to get cold feet over adoption, too. But luckily I came to it through the beta program, so I knew up front this was really NGWS (Next Generation Windows Services), which was vastly more comprehensive than just web services, so I expected a complex and broad class library.
I think you and I agree on the large-scale points though, you just somehow got the impression I was laying the blame on complexity. It's all the fault of really poor marketing, first and foremost. If your developers don't know what it's about, they aren't going to (can't) make very good decisions about if, where, and when to use it.
Unfortunately I can't think of a good way around it. Probably your best bet is to just cough up $40 or $50 for a good book. There are quite a few references springing up now.
Also, unless VB.NET (or older versions of VB) are firmly in place at your shop, I'd seriously recommend looking at C#. It has a lot of the nice features you get from VB such as bounds checking, but it doesn't have the squirrely syntax baggage VB is dragging around from, in some cases, the earliest days of BASIC. BTW, before any VBers whine at me, I was a VB programmer for a long time at one point, too. Hated it, but it paid a lot of big bills.:)
I don't find the hot links half as aggravating as I would if they weren't there.:)
Don't get me wrong, I think their docs have room for improvement, but they're far better than any technical docs Microsoft has produced in the past ten years or so (since their "Quick" line of compilers -- QuickC, QuickBasic, etc).
What I've found about the docs is I can usually drill down to some piece of functionality I need -- even when I've never used it before and I'm not necessarily sure it exists. That says a lot for both the class hierarchy itself and the documentation. Of course, out of thousands of classes there are cases where the correct usage isn't always obvious, but even then you at least have a good starting point for digging up the right details.
It's just a lot bigger than most developers are used to tackling all at once. It's almost like jumping from DOS programming straight into Win32 coding overnight in terms of how much there is to learn, and how much information is out there distracting you while you're looking for something specific.
Actually it's pretty easy to understand, and isn't the slightest bit mythological.
There are three parts: 1. IL/CLR/etc -- this is basically a runtime environment/component model 2. Languages & compilers -- these compile code to IL (and of course we get C#) 3. System object model -- thousands of classes
The web services crap MS marketing is confusing everybody with is just a small subset of the system object model. MS expended great effort defining an extensive and deep model that covers everything from messaging to GUI management. They also made a real effort to isolate MS- and Windows-specific features from more generic features.
Also, it's worth noting that the documentation is actually VERY good. It can be a bit overwhelming because there are tens of thousands of pages, but it's all cross-referenced with hot links, sample code is common, and it's backed up by a large amount of more involved samples.
And all of that is available in the free portion that you can download, the.NET Framework SDK.
It only seems mysterious because Microsoft marketing has fucked up so badly.
Sorry, I don't think your scenario applies here, although it certainly describes a very common problem. This isn't about delivering something late, or failing to deliver on a promise, it's about a hugely incompetent attempt to describe what IS being delivered.
With the exception of the.NET server platforms (which underwent major overhauls not really related to.NET itself),.NET itself was delivered more or less on time, and to people who knew what it was (e.g. we could ignore the marketing) it actually delivered far more than any of us expected.
There is so much to.NET it's difficult for an experienced developer to fully grasp without a long period of study. So it isn't surprising that marketing couldn't boil it down to a nice bite-sized chunklet wrapped in a juicy slogan. (Not that they should try, but sadly that's what seems to sell.)
I can't tell you how happy it makes me to see other people have figured this out. Microsoft marketing might be good at pushing individual products or tightly-coupled suites like Office, but when it comes to selling technologies, they suck badly. When I was beta testing and saw the "services" hype machine gearing up, I wanted to pull my hair out. It's as if a Ferrari salesman chooses to focus on the stereo exclusively. Yes, it's present and useful and some people will be very excited about it, but it ignores a vast array of other more important things, more compelling reasons to invest some effort into this.
During the beta I thought this might be just a smoke screen to keep the DOJ from looking at it too closely. After all, proper exploitation of the CLR should allow them to eventually run Windows on other hardware, or maybe even as a full replacement GUI/pseudo-OS layer on other OSes. However, this stupid murky message has persisted, so now I think it's just marketing incompetence.
Recall that MS marketing almost tanked the previous generation of MS technology with that stupid DNA bullshit. I remember YEARS went by before even many developers understood what DNA actually was -- a set of useful discrete but interoperable products which were related but were not "one big thing".
.NET itself is an excellent move for Microsoft, and since virtually everybody uses Microsoft products, it could eventually be a great thing for Windows users too (although if properly used/implemented, they probably won't know it's being used, which is fine).
I just hope BillG gets his heads out of the clouds long enough to pinpoint the problem, execute the market droids responsible for the mess, and make a cleaner, more digestible push to the people who really need to understand it -- the development community.
Oh yes, and one other point -- the size of the framework may prove to be a sticking point. It's pretty big, so unless you're selling CD-based traditional software, it'll be a hard sell for quite some time. But even the typical/. anti-MS flame-belching troll should at least recognize that MS is smart enough to have accomodated that in their planning.
Yeah, the guy is a total freak (this kind of stuff would even offend most religious people I know) and I hate knowing my thirty bucks is feeding his strange little complex, but UE has one great feature lacking in most of the other larger and complex text editors -- it doesn't crash. I can leave it up for days with 50 or 100 large files open, and it just keeps running. No leaks, no weirdness, and small in-memory footprint.
Also, it had C# language support long before the VStudio.NET beta was stable enough to use regularly.:)
I used NoteTab Pro for a long time, but it would very rarely crash out, and even more rarely, it would change ONE CHARACTER in ONE FILE to some random thing, which would take forever to track down.
Slashdot Mirror
First of all, XM is a Sony operation through-and-through. The main reason XM is doing so well is that Sony has pumped a bunch of money into it, knowing up front it'll run at a loss for a long time. On the other hand, Sirius is an independent business with all the attendant risks and concerns. Unfortunately Sony is a giant, monopolistic bastard of a company who pimps cheap junk at ridiculous prices, so you can guess which side I root for in this battle. :)
Second, there is the satellite problem. Sirius placed a small number of satellites in very high orbit. This means each satellite can cover a relatively wide portion of the Earth's surface. Also, these satellites are in a more stable orbit so they have a longer life expectancy. Launch costs were higher, however. Conversely, Sony opted for low-orbit satellites. This means the launches were cheaper, but the life expectancy is lower, and each satellite has a much smaller area of coverage, meaning Sony had to use significantly larger numbers of satellites.
Worse yet, the XM satellites are unexpectedly having problems holding their orbit, so they're using fuel at roughly 3X the expected rate. This is one of XM's largest problems. Sony faces having to replace the satellites YEARS earlier than expected. I don't remember the exact figures to replace the satellites, but the price tag was vast -- like $700M or $1B or something equally insane.
Third, there is the ground station plan. Both systems use a network of repeater stations. These (combined with receiver onboard buffering) ensure you have a steady signal when you drive through an underpass, for example. Amazingly, Sony only has 75 repeater stations in the US, all near major cities. The Sirius plan calls for 2,500 repeaters spread somewhat evenly throughout the country. I don't recall how many Sirius currently has, but it's already much more than XM.
Fourth, of course, is the question of content. Sirius costs a bit more than XM, but it's also largely commercial-free. Sony charges you a not insigificant amount for XM, and then they sell advertising on almost all their stations anyway.
Fifth is the availability issue. Since XM is a Sony-funded effort, it was available everywhere almost from Day One. However, Sirius has a ramp-up plan which targets major cities first, then rolls out to everyone else over a 12-18 month period. This may be important to frequent travelers.
Sixth and finally, speaking of frequent travelers, another post pointed out that people like truckers are a prime target market for satellite radio. Interestingly, there is another important target market mentioned in the articles, but overlooked here so far (I think) -- ships. Merchant ships and cruise ships are potentially a HUGE market for satellite radio. It turns out that XM was very heavily focused on land-based use, and their offshore coverage is largely accidental. Because of this (and XM's use of low-orbit small-footprint satellites) XM coverage doesn't extend very far into the ocean. However, Sirius took that market into consideration when they planned their service, and their coverage extends quite far out to sea. Again, unfortunately, I forget the exact numbers, but the difference was serious, like 20 miles compared to 400 miles.
It's my opinion that Sirius is the better technology for these reasons and others (even ignoring my intense dislike for Sony), but we all know how often Joe Sixpack pays attention to little details like that.
Again, that was all from memory based on information I read almost a year ago, so please don't beat me up if I got any of it completely wrong (and please correct me). Hope you found it interesting.
One day this error message appeared on my screen:
ERROR: NETBIOS has become Twizzled.
I clicked Ok, and everything continued working normally. I searched the hard drive and couldn't find that text anywhere on the machine. I contacted Artisoft (we were one of their larger customers) and they insisted it wasn't from LANtastic. So I contacted Microsoft (back when they had support people you could talk to) and they also denied complicity.
To this day I'm not exactly sure what "twizzled" means. Never saw the error again.
Subarban?!
These are college students?
"Not a flame, just an observation." :)
I see I have to get nit-picky too.
My point definitely was not that "Einstein showed something".
The point I made was based on accepted theory which SUGGESTS (happy?) something may travel at speeds in excess of the speed of light, and that only acceleration to and/or beyond the speed of light is impossible because the energy required would be infinite.
The original article contains an assumption on the author's part that the theory requires that the speed of light is an inviolable upper boundary. My point was simply that this is an inaccurate interpretation.
Therefore, my point is not invalid. The worst you can say is that my choice of a single word was unfortunate. As another responder noted, perhaps this subject requires nit-picking to a degree which makes such distinctions important, but somehow I suspect my /. response isn't going to influence the researchers involved.
So, as I noted previously, I concede that Einstein didn't SHOW anything in the sense that SHOW can be interpreted as providing incontrovertible evidence (and I suppose I could argue that point, too, if I didn't have a life to get back to). Nonetheless, accepted theory does allow for travel beyond the speed of light, and therefore the original article text's assumption that the speed of light was an absolute upper limit is based on an incorrect interpretation of the applicable theory.
Hopefully you are happier with this over-anal-yzed wording.
You're correct, which doesn't invalidate the point I made, does it? That makes the value of your contribution... well, let's just say I wish the moderation system included "whiny anal nitpicky bastard" as an option.
All Einstein showed was that nothing could accelerate beyond the speed of light. Nothing prevents something from moving beyond the speed of light provided it always moves faster than light.
No, what's truly amazing is that many of those managers don't realize what a fraud that is. After many years in Corporate AMerica, I've come to the stunning conclusion that those kinds of windbags generally think they're actually doing a good job.
Pathetic.
How many times have you cracked open a case to discover that every drive was held in by a single screw? Or that screw that the owner has stripped and is relying on it being wedged in just right to hold the $400 video card in place?
Of course, it's easiest to acquire these items by simply collecting them over time. That would be a cool thing for an online retailer to sell. "500 piece miscellaneous computer fastner pack... $7..." :)
Oh yeah, in case nobody said it: zip-ties and velcro tie-wraps...
Yes, I worded that part very poorly. My point was that it has many of the same attractions that VB has -- easy to learn, easy to deploy, etc. No wild pointers bringing down the service... one of those languages where anybody can start cranking out code without knowing much about how stuff actually works. (I'm not saying that's a good thing.)
I agree that HTML has killed VB on the client, too, but that's not really relevant to the discussion. Indeed, there's probably a larger market for COBOL guys now than VB guys (a guess, but my company sure has a shortage).
Java on the client was mainly killed by plain old HTML and round-trip processing (CGI, ASP, forms, etc), which proved far superior for quickly and easily building decent, usable UIs, and which downloaded to and rendered on end-user machines much more quickly.
The really stupid thing is that Microsoft was probably Sun's best chance for keeping Java alive on the client. MS had the fastest JVM around for a LONG time, and they had the most bug-free JVM for a long time. (I wrote Java apps for a number of years and had to constantly test on about a dozen JVM/JITs, I even recall the first Sun Java event where Sun people were praising and recommending the Microsoft JVM.)
Microsoft probably could have killed Java, and it's reasonable to assume they thought about it -- although I personally disagree that J++ was an actual attempt to do so (it was a kick-ass environment, and I produced vast amounts of full-compatible code with it) -- but the simple fact is they didn't have to, because Java never really got going on the client. Java lives on in the server-side world as a kind of VB-for-people-who-hate-Microsoft, and that's about it, but MS didn't kill it.
I was always worried more about the Gelatinous Cubes, oh yeah, and the Grey Ooze...
03/2001 photonics.com article
The use of apostrophes is an obvious improvement over the quotes used by BASIC: PRINT "Hello, Python." Now THAT's what I call unacceptable syntax. (Of course I know the difference, it just struck me as amusing.)
You also neglect this technical problem in XP: "If you say no to some of the requests, some functions of Windows XP will not work (such as networking)." If you deny internet access to many components, XP will cease to function properly. Did you notice the long list he had of components that needed 'net access? Windows Media Player!?! That's a technical flaw that's borderline malicious.
Untrue. I agree that a stupidly huge number of apps and processes attempt to access the 'net, but it isn't true that XP will cease to function properly if you deny that access. My wife's external firewall is configured to automatically deny access to everything except to a small handful of specific apps/ports/addresses, and XP runs just fine. Yes, including Media Player.
There is also no technical solution things like "Run DLL as an app" not telling you which DLL needs to be run. These programs should not be calling home unless they need to in order to function properly for the user's benefit. They way they work now is simply frivolous.
I do agree with you here in principle, although it isn't technically impossible to determine what DLL is making the call, just difficult. Many of the tools from SysInternals will show you all the nitty-gritty details of what DLLs are in use by each process, for example. Interestingly, Tiny Software's Personal Firewall is still able to block multiple apps using RUNDLL or the generic "service" process by maintaining hashes as identifiers. Unfortunately it's still up to the user to track down what specifically is trying to make the connection, though.
Also it should be noted that XP really doesn't "call home" as you said. At least, not in the sense that you wish to imply. If you pay attention to where the calls go, they do perform some useful task in almost all cases. (The value of these tasks is a completely separate question.) The real problem is that the user isn't given any choice in the matter, or even told that it's happening. But none of the connections I've seen or that I'm aware of were ever specifically a "call home" facility. Yes they might be tracked that way, but they aren't raw "call home" connections as you imply.
If the original poster isn't just lying or wasn't trippin' balls, I'd guess he saw a B2. I've seen B2's pass overhead at night, and they're exactly like what the article describes -- enormous black triangles that make no noise and appear to be moving very slowly (since they're frigging enormous).
I've also seen B2's pass from the side at airshows, and like all very large aircraft, they appear to be moving impossibly slowly. It's an optical illusion, of course, as they have stall speeds well over 100 MPH (a B2 is probably around 115 MPH). Having grown up on military bases I was very accustomed to staring at 100-ton C5 Galaxy transports coming in for landings at what appeared to be liesurely single-digit speeds.
This whole giant secret blimp thing sounds pretty silly, though.
Boy are YOU in the wrong place.
Besides, I didn't say VB support wasn't important, I said it wasn't the definitive litmus test everybody says it is for early adoption of .NET. Remember, the main topic is why .NET is being adopted more slowly than MS expected. If all the VBers in the world made the switch overnight, the sheer size of the Framework (over 100MB) would prevent them from forcing a rapid wide-scale deployment. Many VBers have difficulty making the case for the much, much smaller runtimes in previous versions (what's VB6? something like 6MB now?).
Heck, the machine I'm using now has a whole slew of .NET-based services (in the NT Service sense of the word) running, I've already written two fairly significant .NET applications which aren't web based at all, and I'm in the middle of a third which has the ability to communicate via TCP/IP but definitely isn't web-centric.
Sorry man, but I think you really missed the boat on that one.
While I agree the VB compatability question does play a role, I don't think it's as important as you suggest. With each rev of VB, the previous version (sometimes the previous two versions) remained in widespread use for quite a long time. This has become normal in the VB community.
I believe the actual problem is that Microsoft and particularly Microsoft marketing has done a really poor job of explaining to developers what .NET really is and what it has to offer. Most developers simply don't know what .NET really is all about. I've been to MANY .NET-related shows, seminars, conferences, and other things over the past couple of years and even near the end of the show, people are wandering around confused -- often because what they're hearing has NOTHING (or very little) to do with what the marketing droids have told them to think about.
The complexity is really a side-effect problem. If I thought .NET was only web-services-related, which would be an easy conclusion to draw based on the marketing, and I was then faced with the thousands of classes which comprise .NET, I'd start to get cold feet over adoption, too. But luckily I came to it through the beta program, so I knew up front this was really NGWS (Next Generation Windows Services), which was vastly more comprehensive than just web services, so I expected a complex and broad class library.
I think you and I agree on the large-scale points though, you just somehow got the impression I was laying the blame on complexity. It's all the fault of really poor marketing, first and foremost. If your developers don't know what it's about, they aren't going to (can't) make very good decisions about if, where, and when to use it.
Also, unless VB.NET (or older versions of VB) are firmly in place at your shop, I'd seriously recommend looking at C#. It has a lot of the nice features you get from VB such as bounds checking, but it doesn't have the squirrely syntax baggage VB is dragging around from, in some cases, the earliest days of BASIC. BTW, before any VBers whine at me, I was a VB programmer for a long time at one point, too. Hated it, but it paid a lot of big bills. :)
Don't get me wrong, I think their docs have room for improvement, but they're far better than any technical docs Microsoft has produced in the past ten years or so (since their "Quick" line of compilers -- QuickC, QuickBasic, etc).
What I've found about the docs is I can usually drill down to some piece of functionality I need -- even when I've never used it before and I'm not necessarily sure it exists. That says a lot for both the class hierarchy itself and the documentation. Of course, out of thousands of classes there are cases where the correct usage isn't always obvious, but even then you at least have a good starting point for digging up the right details.
It's just a lot bigger than most developers are used to tackling all at once. It's almost like jumping from DOS programming straight into Win32 coding overnight in terms of how much there is to learn, and how much information is out there distracting you while you're looking for something specific.
There are three parts:
1. IL/CLR/etc -- this is basically a runtime environment/component model
2. Languages & compilers -- these compile code to IL (and of course we get C#)
3. System object model -- thousands of classes
The web services crap MS marketing is confusing everybody with is just a small subset of the system object model. MS expended great effort defining an extensive and deep model that covers everything from messaging to GUI management. They also made a real effort to isolate MS- and Windows-specific features from more generic features.
Also, it's worth noting that the documentation is actually VERY good. It can be a bit overwhelming because there are tens of thousands of pages, but it's all cross-referenced with hot links, sample code is common, and it's backed up by a large amount of more involved samples.
And all of that is available in the free portion that you can download, the .NET Framework SDK.
It only seems mysterious because Microsoft marketing has fucked up so badly.
With the exception of the .NET server platforms (which underwent major overhauls not really related to .NET itself), .NET itself was delivered more or less on time, and to people who knew what it was (e.g. we could ignore the marketing) it actually delivered far more than any of us expected.
There is so much to .NET it's difficult for an experienced developer to fully grasp without a long period of study. So it isn't surprising that marketing couldn't boil it down to a nice bite-sized chunklet wrapped in a juicy slogan. (Not that they should try, but sadly that's what seems to sell.)
During the beta I thought this might be just a smoke screen to keep the DOJ from looking at it too closely. After all, proper exploitation of the CLR should allow them to eventually run Windows on other hardware, or maybe even as a full replacement GUI/pseudo-OS layer on other OSes. However, this stupid murky message has persisted, so now I think it's just marketing incompetence.
Recall that MS marketing almost tanked the previous generation of MS technology with that stupid DNA bullshit. I remember YEARS went by before even many developers understood what DNA actually was -- a set of useful discrete but interoperable products which were related but were not "one big thing".
I just hope BillG gets his heads out of the clouds long enough to pinpoint the problem, execute the market droids responsible for the mess, and make a cleaner, more digestible push to the people who really need to understand it -- the development community.
Oh yes, and one other point -- the size of the framework may prove to be a sticking point. It's pretty big, so unless you're selling CD-based traditional software, it'll be a hard sell for quite some time. But even the typical /. anti-MS flame-belching troll should at least recognize that MS is smart enough to have accomodated that in their planning.
Also, it had C# language support long before the VStudio.NET beta was stable enough to use regularly. :)
I used NoteTab Pro for a long time, but it would very rarely crash out, and even more rarely, it would change ONE CHARACTER in ONE FILE to some random thing, which would take forever to track down.
UE is definitely kick-ass.