These aren't JVMs, they are application servers. *Very* different things, really.
Geronimo, JOnAS, and JBoss all work with Sun's classpath. They may also work with the open source Classpath stuff. But these aren't intended as JVM replacements.
I always laugh at these sort of anecdotes. They only go to show you that a good thinker without a degree in CS is better than an average thinker with a degree in CS.
In my 16 years as a professional programmer, I've had the pleasure of working with 6 different people I would consider to be "greats" at computer programming. The funny thing? All of them had CS degrees. Interestingly, 5 of the 6 went to purely theoretical schools: CSU, Carnegie Mellon, and Rice University (the sixth was educated in Scotland at a university with which I am unfamilar). The interesting thing about this collection of schools? None of them teach programming. They don't teach how to program in a specific language, or to work with a specific operating system. They teach computer science theory. The application of the theory is left up to the student.
I've worked with tons of people that can produce prodigious amounts of code. However, the best programmers I've worked with produce less code, but code that does more. I've met lots of people considered to be "great" programmers by the companies I've worked for. Most produced tons of code--but very few produced well engineered code. This is the difference between a prolific programmer and a good programmer. These so-called "great" programmers are often nothing more than hackers (in the coding sense)--they can whip out some code that meets the purpose in less time than an engineer. But that code is often unmaintainable, inflexible, and has to be rewritten in a few years.
In my 16 years on the job, I've gotten progressively less prolific each year, but the quality of my code has gone up. The better I get, the less code I write, but the more my code is able to do.
I've worked peripherally with many excellent programmers, a number of which had no formal computer science education (and I'm not trying to assert that a CS degree is required to be a "great" programmer). I believe that a good thinker is far more important than one with specific experience or a degree in computer science. Often, however, the best software architects and engineers I've worked with did have excellent educations from purely academic schools. An excellent CS program won't teach you a technology, but it will teach you how to think. It's a shame most universities place so much emphasis on the application of technology rather than on the theory behind the technology.
In addition to using the tag, which is available only to IE users, you can also use
tags and issuing requests to a hidden iframe that posts the results back to the parent window. Using the div tag approach, of course, still requires an iframe, but at least it's cross platform.
Moore's Law states that transister density will double every 18 months. As a corollary, the cost per transister will halve every 18 months. But Moore's Law never promised any specific performance improvements.
Of course, smaller transisters lead to shorter circuit paths (i.e. higher speed). Furthermore, higher transister density leads to greater power consumption, leading manufacturers to work on lower power requirements for transisters (this leads to faster switching times for transisters, and therefore higher speeds). In other words, it can be inferred from Moore's Law that speed will increase. But Moore's Law says nothing about predicting speed increases from circuits.
Does this mean that Rockefeller and Carnegie aren't worthy of being honored because they engaged in predatory business practices, too? From what I remember from my history classes, Rockefeller was particularly nasty.
You can add many more of the great philanthropists of our time in the same category. Usually, to give away a great deal of money one must first make the money. Predatory business practices are one of the most common ways for the extremely rich to make their money. Carrying this to an extreme to make a point, without predatory business practices the world wouldn't get to experience the philanthropy.
It has nothing to do with the keyboard. The problem you're experiencing probably has more to do with your mouse than anything. Most people place the mouse too high and too far away from their body than they should. Mouse clicking with an extended arm is a very common cause of RSI--the motion causes a great deal of strain to the wrist. This is why some people with RSI use a foot pedal as a mouse clicker. In fact, these foot pedal mice buttons are used very commonly in the vector digitization industry (people that trace maps and oil well logs to vectorize raster or image-based data).
My recommendation? Get a track ball. Or get a keyboard tray wide enough to put your mouse on. O
And Ted Turner never would have donated the money unless he met Jane Fonda. What's your point? I don't get it. Last I heard, it was Bill Gates' money that funded the "Bill and Melinda Gates Foundation". See, that's why the Foundation bears his name. Just because he isn't involved in the day-to-day operations of the Foundation doesn't make him any less generous. This kind of reminds me of the addage, "Don't look a gift horse in the mouth."
History is filled with nasty, cutthroat, and very wealthy individuals making huge fortunes by the time they are 40 (without any donations at all), then donating everything before dieing (Rockefeller and Carnegie come to mind). I, for one, welcome any and all money Bill donates, and I look forward to seeing him donate the rest of his fortune before his death. I just hope the current administration doesn't change the inheritance laws such that extremely wealthy individuals no longer have a compelling reason to donate their money to anyone other than their children. Given Bill Gates' and his father's statements on the positive reasons for inheritance taxes, I doubt a change in the laws would affect Bill Gates' generosity. But it would be a shame to see this cycle of wealth redistribution cut off as a result of short-sighted politics.
Black on yellow is the most likely to catch our attention, particularly in our peripheral vision. But the eye is most sensitive to green. I wouldn't recommend black-on-green, though.
Higher contrast may be harder on the eyes, but I'm not sure how that matters when viewing a laptop in sunlight. The contrast will be so poor if the bright bits of the screen are illuminated by a backlight as opposed to the sunlight. Try getting one without a backlight at all (like the *original* mac laptops from a *long* time ago). I don't know of any, but I'm not up with tablet PCs, either.
Whatever you do, don't wear polarized sunglasses (unless you decide to mod your laptop by removing its polarized glass). In order to display data on an LCD monitor the backlight is passed through polarized glass on the laptop screen. Wearing polarized sunglasses will only cause the wearer to tilt their head at a funny angle so that they can see the screen at all. From my experience, the polarization of LCD monitors is usually at a 45 degree angle to the polarization of most sunglasses. If you do decide to mod your laptop, you'll end up with a screen that appears blank to most people, though, which is quite an interesting effect. But if you hold your head at the wrong angle the entire screen goes black. Not good.
Start with free email. The free email provides a number of features. Start with disposable email addresses. These are email aliases where you can get an expiry date. Anything received after the expiry date at the disposable email address gets forwarded to/dev/null. Anything before then gets forwarded to your "real" email address. And, since they're the Gods at Google, they'll give us tons of free space...maybe 20MB...disk space is cheap these days. All with a unique and easy-to-remember email address. How about @googlemail.com. Couple that with Bayesian filtering to get rid of spam.
Then add a few extra services for pay. Extra space. Full automatic encryption with a shared public key repository all using GPG. An anonymizer plug-in that mates with the disposable email address to create anonymous web browsing using a false 'nic'. Whatever else those guys can think of, too. Charge something reasonable for each additional feature...something like $10-20/year for each. I could see a day when the average broadband user pays $30/mo for their network service and $10/mo for additional google services. But the poor college student still gets a good deal for free. And Google gets those customers whether they buy their networking services from Time Warner, SBC, or AOL. That could justify Google's massive net worth when they go public.
I'd pay for some of those services. I trust Google. Combine all that with Google's reliable follow-through on all their endeavours and I bet Google, if anyone, can deliver.
What major flavor of Unix does Java not run on? AIX? Check. Solaris? Of course. Irix? Check. MacOSX? Check. PPC Linux? Check. BSD? Check.
As for the non-free bit, Java runs on MacOS X, Irix, AIX, and Windows. I'm sure there are more ports, but that covers all of the major OSes that run server-side software. J2ME even runs on tons of telephones and PDAs.
I'm not sure what your point is. But it seems like your information may be a bit out of date.
I hope this is the caliber of work I can expect from CD in the future. These are the folks that left the JBoss Group after the incident concerning Apache's launch of the Geronimo J2EE server project (do a google search on "elba jboss geronimo" if you're interested in learning more about the incident). This is an excellent article on a very timely subject. I wish all new technical specifications had such a thorough analysis of their impact upon their release. Thanks, Greg Wilkins!
After losing a pair of ER-4 (the original ones), I just bought a pair of ER-6. Now, I don't have the two to compare side-by-side, but I'd swear to you the ER-6 sounds closer to my reference (a pair of Stax Lambda Pros) than the ER-4. At 1/3 the price, with better sound isolation and better comfort on the ears, I'll take the ER-6 any day.
By the way, nothing comes close to the Stax. Despite the reviews I've read on the Etymotic Research (and just about every other audio transducer I've read about), nothing comes close to the Stax. I love my Stax.
Wouldn't it be possible to store fonts in an SVG format, then render each character to a buffer for a given point size? In that sense, wouldn't SVG be capable of replacing TTF (TrueType Font) file format? I wouldnt' think it would take much logic to render the longest straight vertical edge to align exactly to a hard pixel break, then anti-alias all edges that don't align to a pixel boundary. I haven't ever seen a TTF renderer, but my guess is that is all that happens--TTF is a vector format, and when a TTF font is loaded and used at a given point size, the renderer aligns the bottoms and longest straight verticle edge of characters to a hard pixel break. Anti-aliasing should be able to performed on the video card as long as the font renderer renders anti-aliased pixels with a transparency component. Such a mythical SVG font renderer would have reasonable performance and be able to take advantage of hardware acceleration for anti-aliasing. Map the pixel-rendered output to texture memory and you should be able to take advantage of fancy screen effects used in MacOSX and Windows XP Media Edition.
It's just an old 50" rear projection big screen...no HDTV, but a good enough picutre for the TiVo. I don't really use the VCR any more. But I do use the DVD. Perhaps DiVX will be in my future....
When comparing football fields, it's imortant to compare apples to apples. The length of the Canadian football field is 110 yards (which, although close, is not 100m). The endzones on a Canadian football field are 20 yards deep (each). An American football field is 100 yards in length. The endzones are each 10 yards deep.
Despite the erroneous quote you found that compared the length of one field to another, the relative lengths of the fields are either 110 yds vs. 100 yds (Canadian vs. American using just the playing field) or 150 yds vs. 120 yds (Canadian vs. American using the playing filed plus the end zones).
Here's my experiences working with the latest OpenOffice RC (I believe that is what StarOffice 7 is based upon):
Writer is pretty good, but has some serious flaws with page numbering. Namely, there is no concept with OO/SO of sections in the same way as MS Word. You have to bend over backwords to make it break a document so that the table of contents, for example, is numbered using lower-case roman numerals while the main body is numbered starting from 1 using Arabic numerals. Creating a document that excludes the page number from the first page but prints it on all other pages is also a pain in the ass. Importing MS Word documents that are set up this way is broken. Changing formats for heading styles half way through a document is also broken. Resetting numbering for outlines half way through a document is also broken. Every complex document I've ever worked on utilizes all of these features. OpenOffice is very nice, but these features are a necessity for me. In my opinion, this makes OpenOffice unusable for complex documents, and makes its use for interoperability somewhat limited (although interoperability is less likely an issue when dealing with complex documents).
Calc is very good, and I have only noticed a single obscure problem. Excel allows spreadsheets with 65,565 rows, while Calc only allows spreadsheets with 32,767 rows. This is an obscure limit, and I would recommend against creating any spreadsheet that pushes this limit. However, if converting an entire organization to OpenOffice/StarOffice, this may be a problem. A bigger problem for conversion would likely be the lack of Visual Basic support. I don't consider the row size limit to be a show stopper (whereas the Writer limitations are show stoppers, imho). Calc is very good as a whole. However, if your organization relies heavily upon VB macros, then you should consider the effects this will have on any migration.
I have not stress-tested Impress enough to notice any limitations/bugs. So far, everything I have thrown at Impress comes through fine.
As for formatting, I have only had minor issues regarding formatting (like a single line being thrown onto the next page with a document). These issues are similar to those encountered when changing printer types under MS Word. They are a nuiscance, but not a show stopper.
This "comprehensive" review was anything but. If the author had investigated OO/SO's shortcomings even a little bit, the page numbering issues would have been apparent. But, hey, that's what Slashdot is for, no?
Mine died after 13 months (read: 1 month after expiry of warranty). My brother has gone through 3 Treos. I hope this one is more sturdy (or is that sturdier?).
While MS may do everything in their power to establish and control proprietary de facto standards, their work on XML and XSL consistently have pushed standards. In case you don't remember, MS owns a patent or two covering the use of stylesheets, and specifically the technology behind XSLT. They have stated that they will not enforce these patents whatsoever. They have been the driving force behind SOAP, with IBM and others joining in on the bandwagon after they determined SOAP has strong potential.
They may have applied XML in a very non-standard way with MS Office, but their work on the standards themselves (XML, XSL, SOAP, and related tech) is quite exemplary.
Actually, just put StarOffice and MadHatter on your desktops, and you've got a supported corporate solution for both desktop & server software for $100 per employee. You still need to buy the hardware. But I think that's pretty competitive, all things considered. You can't get an MS-based solution consisting of MS Office, Windows, Exchange, and.NET for anywhere near that price.
Yes. Perhaps it was. That is why the selection of a control group is necessary. I'm really suggesting that this warrants further study. I wish the US military at least would study this issue more closely. To date, the DU studies have been very incomplete, looking only at exposure to external DU instead of looking at the effects of ingested DU.
Slashdot Mirror
These aren't JVMs, they are application servers. *Very* different things, really.
Geronimo, JOnAS, and JBoss all work with Sun's classpath. They may also work with the open source Classpath stuff. But these aren't intended as JVM replacements.
I always laugh at these sort of anecdotes. They only go to show you that a good thinker without a degree in CS is better than an average thinker with a degree in CS.
In my 16 years as a professional programmer, I've had the pleasure of working with 6 different people I would consider to be "greats" at computer programming. The funny thing? All of them had CS degrees. Interestingly, 5 of the 6 went to purely theoretical schools: CSU, Carnegie Mellon, and Rice University (the sixth was educated in Scotland at a university with which I am unfamilar). The interesting thing about this collection of schools? None of them teach programming. They don't teach how to program in a specific language, or to work with a specific operating system. They teach computer science theory. The application of the theory is left up to the student.
I've worked with tons of people that can produce prodigious amounts of code. However, the best programmers I've worked with produce less code, but code that does more. I've met lots of people considered to be "great" programmers by the companies I've worked for. Most produced tons of code--but very few produced well engineered code. This is the difference between a prolific programmer and a good programmer. These so-called "great" programmers are often nothing more than hackers (in the coding sense)--they can whip out some code that meets the purpose in less time than an engineer. But that code is often unmaintainable, inflexible, and has to be rewritten in a few years.
In my 16 years on the job, I've gotten progressively less prolific each year, but the quality of my code has gone up. The better I get, the less code I write, but the more my code is able to do.
I've worked peripherally with many excellent programmers, a number of which had no formal computer science education (and I'm not trying to assert that a CS degree is required to be a "great" programmer). I believe that a good thinker is far more important than one with specific experience or a degree in computer science. Often, however, the best software architects and engineers I've worked with did have excellent educations from purely academic schools. An excellent CS program won't teach you a technology, but it will teach you how to think. It's a shame most universities place so much emphasis on the application of technology rather than on the theory behind the technology.
Your sig is hillarious. I wonder how many people that didn't study CS in university will get it.
If this were true, then why did CS enrollment at universities drop between 1985 and 1995?
In addition to using the tag, which is available only to IE users, you can also use tags and issuing requests to a hidden iframe that posts the results back to the parent window. Using the div tag approach, of course, still requires an iframe, but at least it's cross platform.
Moore's Law states that transister density will double every 18 months. As a corollary, the cost per transister will halve every 18 months. But Moore's Law never promised any specific performance improvements.
Of course, smaller transisters lead to shorter circuit paths (i.e. higher speed). Furthermore, higher transister density leads to greater power consumption, leading manufacturers to work on lower power requirements for transisters (this leads to faster switching times for transisters, and therefore higher speeds). In other words, it can be inferred from Moore's Law that speed will increase. But Moore's Law says nothing about predicting speed increases from circuits.
Does this mean that Rockefeller and Carnegie aren't worthy of being honored because they engaged in predatory business practices, too? From what I remember from my history classes, Rockefeller was particularly nasty.
You can add many more of the great philanthropists of our time in the same category. Usually, to give away a great deal of money one must first make the money. Predatory business practices are one of the most common ways for the extremely rich to make their money. Carrying this to an extreme to make a point, without predatory business practices the world wouldn't get to experience the philanthropy.
It has nothing to do with the keyboard. The problem you're experiencing probably has more to do with your mouse than anything. Most people place the mouse too high and too far away from their body than they should. Mouse clicking with an extended arm is a very common cause of RSI--the motion causes a great deal of strain to the wrist. This is why some people with RSI use a foot pedal as a mouse clicker. In fact, these foot pedal mice buttons are used very commonly in the vector digitization industry (people that trace maps and oil well logs to vectorize raster or image-based data).
My recommendation? Get a track ball. Or get a keyboard tray wide enough to put your mouse on. O
And Ted Turner never would have donated the money unless he met Jane Fonda. What's your point? I don't get it. Last I heard, it was Bill Gates' money that funded the "Bill and Melinda Gates Foundation". See, that's why the Foundation bears his name. Just because he isn't involved in the day-to-day operations of the Foundation doesn't make him any less generous. This kind of reminds me of the addage, "Don't look a gift horse in the mouth."
History is filled with nasty, cutthroat, and very wealthy individuals making huge fortunes by the time they are 40 (without any donations at all), then donating everything before dieing (Rockefeller and Carnegie come to mind). I, for one, welcome any and all money Bill donates, and I look forward to seeing him donate the rest of his fortune before his death. I just hope the current administration doesn't change the inheritance laws such that extremely wealthy individuals no longer have a compelling reason to donate their money to anyone other than their children. Given Bill Gates' and his father's statements on the positive reasons for inheritance taxes, I doubt a change in the laws would affect Bill Gates' generosity. But it would be a shame to see this cycle of wealth redistribution cut off as a result of short-sighted politics.
I see a few issues with this advice.
Black on yellow is the most likely to catch our attention, particularly in our peripheral vision. But the eye is most sensitive to green. I wouldn't recommend black-on-green, though.
Higher contrast may be harder on the eyes, but I'm not sure how that matters when viewing a laptop in sunlight. The contrast will be so poor if the bright bits of the screen are illuminated by a backlight as opposed to the sunlight. Try getting one without a backlight at all (like the *original* mac laptops from a *long* time ago). I don't know of any, but I'm not up with tablet PCs, either.
Whatever you do, don't wear polarized sunglasses (unless you decide to mod your laptop by removing its polarized glass). In order to display data on an LCD monitor the backlight is passed through polarized glass on the laptop screen. Wearing polarized sunglasses will only cause the wearer to tilt their head at a funny angle so that they can see the screen at all. From my experience, the polarization of LCD monitors is usually at a 45 degree angle to the polarization of most sunglasses. If you do decide to mod your laptop, you'll end up with a screen that appears blank to most people, though, which is quite an interesting effect. But if you hold your head at the wrong angle the entire screen goes black. Not good.
Here's what I'd love to see.
/dev/null. Anything before then gets forwarded to your "real" email address. And, since they're the Gods at Google, they'll give us tons of free space...maybe 20MB...disk space is cheap these days. All with a unique and easy-to-remember email address. How about @googlemail.com. Couple that with Bayesian filtering to get rid of spam.
Start with free email. The free email provides a number of features. Start with disposable email addresses. These are email aliases where you can get an expiry date. Anything received after the expiry date at the disposable email address gets forwarded to
Then add a few extra services for pay. Extra space. Full automatic encryption with a shared public key repository all using GPG. An anonymizer plug-in that mates with the disposable email address to create anonymous web browsing using a false 'nic'. Whatever else those guys can think of, too. Charge something reasonable for each additional feature...something like $10-20/year for each. I could see a day when the average broadband user pays $30/mo for their network service and $10/mo for additional google services. But the poor college student still gets a good deal for free. And Google gets those customers whether they buy their networking services from Time Warner, SBC, or AOL. That could justify Google's massive net worth when they go public.
I'd pay for some of those services. I trust Google. Combine all that with Google's reliable follow-through on all their endeavours and I bet Google, if anyone, can deliver.
Yes, the last time a pilotless aircraft was used in a suicide mission everything went swimmingly!
What major flavor of Unix does Java not run on? AIX? Check. Solaris? Of course. Irix? Check. MacOSX? Check. PPC Linux? Check. BSD? Check.
As for the non-free bit, Java runs on MacOS X, Irix, AIX, and Windows. I'm sure there are more ports, but that covers all of the major OSes that run server-side software. J2ME even runs on tons of telephones and PDAs.
I'm not sure what your point is. But it seems like your information may be a bit out of date.
I hope this is the caliber of work I can expect from CD in the future. These are the folks that left the JBoss Group after the incident concerning Apache's launch of the Geronimo J2EE server project (do a google search on "elba jboss geronimo" if you're interested in learning more about the incident). This is an excellent article on a very timely subject. I wish all new technical specifications had such a thorough analysis of their impact upon their release. Thanks, Greg Wilkins!
After losing a pair of ER-4 (the original ones), I just bought a pair of ER-6. Now, I don't have the two to compare side-by-side, but I'd swear to you the ER-6 sounds closer to my reference (a pair of Stax Lambda Pros) than the ER-4. At 1/3 the price, with better sound isolation and better comfort on the ears, I'll take the ER-6 any day.
By the way, nothing comes close to the Stax. Despite the reviews I've read on the Etymotic Research (and just about every other audio transducer I've read about), nothing comes close to the Stax. I love my Stax.
Wouldn't it be possible to store fonts in an SVG format, then render each character to a buffer for a given point size? In that sense, wouldn't SVG be capable of replacing TTF (TrueType Font) file format? I wouldnt' think it would take much logic to render the longest straight vertical edge to align exactly to a hard pixel break, then anti-alias all edges that don't align to a pixel boundary. I haven't ever seen a TTF renderer, but my guess is that is all that happens--TTF is a vector format, and when a TTF font is loaded and used at a given point size, the renderer aligns the bottoms and longest straight verticle edge of characters to a hard pixel break. Anti-aliasing should be able to performed on the video card as long as the font renderer renders anti-aliased pixels with a transparency component. Such a mythical SVG font renderer would have reasonable performance and be able to take advantage of hardware acceleration for anti-aliasing. Map the pixel-rendered output to texture memory and you should be able to take advantage of fancy screen effects used in MacOSX and Windows XP Media Edition.
It's just an old 50" rear projection big screen...no HDTV, but a good enough picutre for the TiVo. I don't really use the VCR any more. But I do use the DVD. Perhaps DiVX will be in my future....
I've got my TiVo, VCR, & TV on a UPS. That solves the flashing 12:00 problem most of the time.
When comparing football fields, it's imortant to compare apples to apples. The length of the Canadian football field is 110 yards (which, although close, is not 100m). The endzones on a Canadian football field are 20 yards deep (each). An American football field is 100 yards in length. The endzones are each 10 yards deep.
Despite the erroneous quote you found that compared the length of one field to another, the relative lengths of the fields are either 110 yds vs. 100 yds (Canadian vs. American using just the playing field) or 150 yds vs. 120 yds (Canadian vs. American using the playing filed plus the end zones).
Of course, SO has the same problems as OO. Try using the app and you'd be able to tell that.
Here's my experiences working with the latest OpenOffice RC (I believe that is what StarOffice 7 is based upon):
Writer is pretty good, but has some serious flaws with page numbering. Namely, there is no concept with OO/SO of sections in the same way as MS Word. You have to bend over backwords to make it break a document so that the table of contents, for example, is numbered using lower-case roman numerals while the main body is numbered starting from 1 using Arabic numerals. Creating a document that excludes the page number from the first page but prints it on all other pages is also a pain in the ass. Importing MS Word documents that are set up this way is broken. Changing formats for heading styles half way through a document is also broken. Resetting numbering for outlines half way through a document is also broken. Every complex document I've ever worked on utilizes all of these features. OpenOffice is very nice, but these features are a necessity for me. In my opinion, this makes OpenOffice unusable for complex documents, and makes its use for interoperability somewhat limited (although interoperability is less likely an issue when dealing with complex documents).
Calc is very good, and I have only noticed a single obscure problem. Excel allows spreadsheets with 65,565 rows, while Calc only allows spreadsheets with 32,767 rows. This is an obscure limit, and I would recommend against creating any spreadsheet that pushes this limit. However, if converting an entire organization to OpenOffice/StarOffice, this may be a problem. A bigger problem for conversion would likely be the lack of Visual Basic support. I don't consider the row size limit to be a show stopper (whereas the Writer limitations are show stoppers, imho). Calc is very good as a whole. However, if your organization relies heavily upon VB macros, then you should consider the effects this will have on any migration.
I have not stress-tested Impress enough to notice any limitations/bugs. So far, everything I have thrown at Impress comes through fine.
As for formatting, I have only had minor issues regarding formatting (like a single line being thrown onto the next page with a document). These issues are similar to those encountered when changing printer types under MS Word. They are a nuiscance, but not a show stopper.
This "comprehensive" review was anything but. If the author had investigated OO/SO's shortcomings even a little bit, the page numbering issues would have been apparent. But, hey, that's what Slashdot is for, no?
Mine died after 13 months (read: 1 month after expiry of warranty). My brother has gone through 3 Treos. I hope this one is more sturdy (or is that sturdier?).
While MS may do everything in their power to establish and control proprietary de facto standards, their work on XML and XSL consistently have pushed standards. In case you don't remember, MS owns a patent or two covering the use of stylesheets, and specifically the technology behind XSLT. They have stated that they will not enforce these patents whatsoever. They have been the driving force behind SOAP, with IBM and others joining in on the bandwagon after they determined SOAP has strong potential.
They may have applied XML in a very non-standard way with MS Office, but their work on the standards themselves (XML, XSL, SOAP, and related tech) is quite exemplary.
Actually, just put StarOffice and MadHatter on your desktops, and you've got a supported corporate solution for both desktop & server software for $100 per employee. You still need to buy the hardware. But I think that's pretty competitive, all things considered. You can't get an MS-based solution consisting of MS Office, Windows, Exchange, and .NET for anywhere near that price.
Yes. Perhaps it was. That is why the selection of a control group is necessary. I'm really suggesting that this warrants further study. I wish the US military at least would study this issue more closely. To date, the DU studies have been very incomplete, looking only at exposure to external DU instead of looking at the effects of ingested DU.