No, it's because a significant number of Slashdotters are programmers, who are interested in programming and improving the electronic devices they own in order to improve their quality of life, and Linux running on a set-top box is a significant factor in this regard.
Granted, for people who don't understand the technology they purchase this may seem like an odd obsession, but for those of us who do, it has far less to do with zealotry than with practicality. Open systems allow those of us who know how to make the systems that we purchase work for us in the ways we want them to. Simple as that.
The thing is that with AV equipment, the source material (CDs and DVDs) stays the same, but the quality of the delivery changes according to how high-end your hardware is. With games, on the other hand, (console games, at any rate) the titles themselves are hugely different between (e.g.) a PS2 and a PS3. The development costs of making a game that utilizes the full potential of the system are going to be vastly different for the PS3 vs. the PS2, and if the PS3 can't gain market share that will inexorably lead to there being a dearth of content for the system.
I really don't think that a high-end/low-end economy can thrive when developing high-end content would require an entirely separate and far more expensive delivery channel.
Right on. I think the only reason more people don't point out the parallels between HD-DVD vs. Blu-Ray to the last format war (SACD vs. DVD-Audio) was that SACD and DVD-Audio were both such catastrophic failures that only audiophiles even know what they were. Because in the mean time, mp3s took over.
And if VHS was 'technically inferior' to Betamax, mp3 compared to SACD/DVD-A would be off the charts bad. And guess what people chose?
Still, I'm wondering what's going on with Zelda. Getting outsold by Wii Sports and Wii Play either means the series is out of favour, people are waiting for the Cube version, Nintendo didn't make enough discs (are they stupid?) or perhaps even that more casual gamers represent a much bigger demand and most Zelda fans can't find a Wii to buy. You'd expect the Zelda folk to be the ones camping out though.
You get Wii Sports with the console, and you get Wii Play when you buy a second controller. That in and of itself should be enough to explain how they outsell Zelda overall, especially since a lot of non-traditional gamer types are buying these things, many of whom have never heard of Zelda. I imagine very few at-launch buyers who do buy Zelda don't buy a second controller at the same time, so there's no way for Zelda to gain numbers on Wii Play and Wii Sports.
If Zelda doesn't sell a lot more than any non-bundled launch title, though, I will be very surprised.
> As for the '30 game rate to break even' 30 games isn't that many > (Should imagine they're are quite a number of people with 30+ ps2 > games), and don't forget Sony get money on Blu-ray discs too.
I don't know... "quite a number" I will grant you. But more than half? More importantly, all things considered, does the mean work out to 30 or more? I would seriously doubt that.
You have a point on Blu-ray, but it may be a catch-22: Sony intended for the PS3 to rocket Blu-ray to the top as the new standard. If the PS3 fails, though, Blu-ray may not end up meaning much more than UMD.
Subscribers can see the stories before the riff-raff (but posting is only possible after the official publication). This actually works out well for everyone, since it increases the chances that some of the early comments might be actually be on-topic and thought through rather than endless 'first post' mispellings. And subscribers can visit sites before they're slashdotted.
I meant to mod you 'insightful' (since my wife behaves the same way with those two games) but my finger slipped and it hit 'redundant' instead. There is no way to undo a moderation directly, but by posting in this discussion, my moderation will be undone and all will be set right. Please disregard this comment, everybody!
In fairness to GP, any analogy breaks down if you push it far enough, and proxy wars are a well-known phenomenon, of which the Korean war was one example. They don't have to be between superpowers either: witness the US-armed and Iranian-armed conflict we just had in South Lebanon...
Sonic Rush is a great 'classical' 2D platformer, and makes incredible use of the dual screens. Its graphics are good enough that it's one of my 'show off the system' titles, and the game is fun whether you have 2 minutes or half an hour to play.
Just FYI (since your post is totally clear given the context): "Shinobi" isn't the name of the character in the Shinobi series of games--it's a common Japanese word which is a synonym for "ninja". So the correct term would be "the ninja from the Shinobi games"--just like "Contra" isn't the name of the character either, so you'd have to say "the soldier from the Contra games" (which incidently, I would add to your list, along with Alex Kid).
I haven't tried it but have heard good things about WebHuddle, which is actually aimed for business meetings (you can even show slides, although for a whiteboard just leave a blank slide and scribble on that). Being aimed at a non-geek audience, setup is intended to be a no-brainer.
You need to know *exactly* what your server needs to serve.
A Mac Mini sure seems like the coolest possible "suitcase server" imaginable, but if you want or might want commercial "enterprise" products like DB2, Oracle, Sybase, WebLogic or WebSphere, your list of options gets quickly shortened for you. Neither OS X or Linux PPC are going to do the job, because you rely on closed-source software whose 'Linux support' is implicitly x86 only (plus *maybe* x86-64). Wait for the Intel-based mini I suppose, and make sure first that the packages you need will run on it.
But as I imagine others will point out, a 1U server for development work would be huge overkill; a remote server or simply running everything on your development machine (through VMWare if necessary) makes way, WAY more sense here, for a myriad of reasons. On the other hand, if your "server" world is 100% open source, you may well be able to craft the ultimate travel server exactly to your specifications with a geek-cool factor that is off the scale. But you don't *need* to; I've written AS/400 WebSphere apps with only Tomcat on Windows (sadly)... but hey, if that's the kind of work you're doing, good luck fitting an AS/400 into your briefcase...
Very insightful post. Obvious when you think about it, but the kind of thing that doesn't occur to you to think through when you're looking at the box in the store. But that has got to be a big factor in the let-down of multiplayer games in modern settings.
Look at the interactive fiction community--lots of great stuff is getting put out for free, and there's a non-trivial amount of coding involved in putting out an IF game.
But getting from that great story/gameplay to a 3d graphics-and-sound fest that would draw in casual gamers is a *huge* amount of work that would require many developer hours, and very little in the way of artistic creativity.
I think that this parallels a wider trend one sees in Open Source where interesting projects tend to be extremely succesful and dull/monotonous areas tend not to be addressed so well. Creativity is not lacking in open source games, as the IF community shows, but marshalling the resources necessary to make compelling 3d graphical games is beyond the community, due to the monotonous nature of the work.
This has indeed been around for years, but it is an opportunity to point out an advantage of the program which I have benefitted from that may be of interest to other slashdotters: yes, it is an opportunity to get to know VMS, HP/UX, and Tru64 Unix (and a few Linux distros, but you have no excuse for not running those on your own hardware). Granted, given the current shift in direction of HP away from (non-Windows based) server tech it is probably not that valuable as career knowledge to familiarise yourself with these OSes, but is still neat for people interested in exploring different operating systems.
But the Test Drive program also gives you access to compilers for a few languages that are tricky to find or unavailible in free software. IIRC you get Fortran 95 on Unix (which in 2006 may be less of a big deal than it was awhile back when there were no decent Free compilers for fortran, as I understand there has been progress on that front lately in gcc-land), and on VMS (if it's still the case--I whined specifically for this back when Compaq was running the program and they added it) you get a COBOL compiler--which there is still no full-featured Free software version of (yes, probably for good reason). Nonetheless, people interested in learning Fortran 95 or Cobol may find the TestDrive program to be the most accessible way of gaining access to these languages.
(And a third benefit, IIRC a few of the servers have Oracle already set up for you, which may be handy for people wanting to learn Oracle but unwilling/unsuccesful in installing it themselves. Although the argument could be made that managing to install Oracle yourself should be the first step in going about learning it...)
While the margins IBM makes on their i5 and z9 servers are huge, they actually do do a lot of dragging into the future as well. Both architectures are sold under two pricing schemes, one with all the legacy support and green screens enabled, and the other with only modern protocols enabled.
Companies that need the legacy support pay roughly ten times what you pay if you only use the modern (Java, WebSphere and DB2) stuff. So although they provide it, and profit heavily from it, they still also heavily encourage switching to Java and WebSphere rather than COBOL and CICS. They just allow companies to move once the price gap gets to the point that it becomes in their economic interest to do so instead of forcing them. "Dragging your customers into the future" has to be kept very distinct from "chasing them away". But they are turning up the heat more and more through licensing.
You can use Java with GTK, Cocoa, or native Windows toolkits. The question then becomes, why Java programmers are not interested.
The basic reason for this is, your application's GUI becomes completely unportable, and you suddenly have little reason for having written it in Java in the first place instead of the platform's "standard" language (C, Objective C, or C#).
A "middle way" is to do what SWT does and wrap the native widgets with a generic API. This creates a "lowest common denominator" problem, however, since you inevitably have to stick to only using those widgets which all your target platforms have...
The reason we see famine in Africa is that Africa has increased its population by 10 in the last 70-80 years.
Really? 10 isn't that many people for Africa to have increased its population by. Heck, I know of individual families that have ten kids in them, and you never hear of it causing a crisis for their town, let alone a whole continent.
As you evidently did not recognise my nick, it is Russian, which I speak fluently and (in addition, since speaking a language doesn't give one the knowledge of its origins) also know that it is descended from Slavonic, a language I also read, which in turn is also NOT descended from Greek (any more than Germanic, Persian, Italic, Sanskrit, or any other branches of our Indo-European tree). I've also spent an absurd number of years studying classics and am quite literate in Latin and Ancient Greek, and am a language geek generally (if I even needed to point that out at this point).
The tsar/caesar connexion is unrelated to linguistics, as are the reasons why Byzantine Greeks called themselves Romans. Just like Finland is often culturally grouped with the Scandinavian countries even though lingistically Swedish is more closely related to Urdu than Finnish. The kings of England claimed the throne of France for centuries too, and yet neither that nor the centuries of Norman rule before that could erase English's Germanic character and turn it into a Romance language.
You are correct that Byzantine civilisation had a lot to do with cementing the cultural influence of Greek, but the Indians were already speaking an Indo-European language for centuries before Alexander showed up there, as were the inhabitants of the European territories of the empire.
You are also correct that a huge amount of the scientific and medical terminology that people often call "Latin roots" are actually Greek roots, but such terms are neologisms anyway.
For fun I (unscientifically and hastily) counted the etymologies up in the last paragraph and got the following numbers (not counting duplicate words, and splitting "terminology" in half!):
Germanic: 14 Latin: 9.5 Greek: 1.5
* Only 'neologism' was fully Greek, ironically. Although the very next paragraph brought out "etymology" and "paragraph", but even so, not that many English words have Greek origins.
I'm sure wikipedia is a great resource to read up on a lot of this stuff, they have an article on "Historical linguistics".
Well, just to back up what your first corrector was saying in a little less flamey way, you are really, really wrong about that.
1. Most of English is not German. English is a Germanic language (as is German) but this ancestry does not account for most of our vocabulary. I read somewhere that only about 500 words of modern English can actually be traced back to Anglo-Saxon (though they are important words, and Anglo-Saxon had a much bigger influence on English grammar than the other languages we get our vocabulary from.
Analogy: My last name in English, but my blood is only 1/8 English since most of my (female) ancestors came from elsewhere. English is a Germanic language in the same sense.
2. Most of English vocabulary (>50%) is LATIN in origin, and most (but not all) of those words came into the language through FRENCH, which is itself descended from Latin.
3. Not only is German not a descendent of Greek, but neither is French, nor Latin, nor in fact ANY language other than modern Greek. This is what makes your assertion so funny to someone who has studied linguistics, because Greek is a very isolated branch of the Indo-European language family (although Greek culture being the foundation of Western Civilisation, there are many vocabulary words in all Western languages that are Greek borrowings. But no languages other than modern Greek are descended from Ancient Greek.)
An actual true statement is that English is a Germanic language and most of its vocabulary comes from Latin (through Norman French). (Although "most" in the previous sentence must mean in terms of number of words, not number of words weighted by frequency of usage.)
I can understand how someone might mix up German and Germanic (different languages, different eras, the similarity of names notwithstanding) and Latin and Greek (as pillars of our classical tradition), but the end result is spectacularly wrong, on so many levels.
I recommend reading up on linguistic history, it's a fascinating field and teaches you a lot about our world, our history, and our species.
1. OAS is not the leading product in the J2EE market either, and I would guess it's actually less popular than JBoss.
2. Total agreement about the oversaturation. IBM seems to be the dominant player all the same, however, and look what they're just starting to do after having bought Gluecode (a company that was putting out an easy-to-use install of the open-source Apache Geronimo): release it as "Websphere Community Edition", boosting their brand awareness at little cost to the company, and gearing up to provide well-supported migration to the 'big' WebSphere.
That strategy seems like a good one for building up market share in a saturated market, and Oracle might just have decided to copy it.
Obviously anyone who prefers modern Gnome or sees the reasoning behind it evolving the way it has and agrees with it should use Gnome. Someone who liked Gnome pre-2.2, when it had a much geekier feel (in terms of using Sawfish, being hackable in so many programming languages, etc.), faces a choice at some point. Either militate and try to gather support for pushing Gnome back in the direction it came from (an unattractive option since major companies like Novell and Sun will be pushing back, and it would seem to involve advocating moving the project backwards rather than forward), or else jump ship and switch to KDE or a window manager that suits his taste.
I would advocate the second option, both to save the Gnome project from unproductive whiners and allow it to make progress that is sure to be very beneficial to business deployments of GNU/Linux desktops, and to allow this hypothetical "ordinary geek" to be happier elsewhere.
Slashdot Mirror
By that logic (just as equal, if not more important, no need to differentiate), they should also be calling themselves men.
No, it's because a significant number of Slashdotters are programmers, who are interested in programming and improving the electronic devices they own in order to improve their quality of life, and Linux running on a set-top box is a significant factor in this regard.
Granted, for people who don't understand the technology they purchase this may seem like an odd obsession, but for those of us who do, it has far less to do with zealotry than with practicality. Open systems allow those of us who know how to make the systems that we purchase work for us in the ways we want them to. Simple as that.
The thing is that with AV equipment, the source material (CDs and DVDs) stays the same, but the quality of the delivery changes according to how high-end your hardware is. With games, on the other hand, (console games, at any rate) the titles themselves are hugely different between (e.g.) a PS2 and a PS3. The development costs of making a game that utilizes the full potential of the system are going to be vastly different for the PS3 vs. the PS2, and if the PS3 can't gain market share that will inexorably lead to there being a dearth of content for the system.
I really don't think that a high-end/low-end economy can thrive when developing high-end content would require an entirely separate and far more expensive delivery channel.
Right on. I think the only reason more people don't point out the parallels between HD-DVD vs. Blu-Ray to the last format war (SACD vs. DVD-Audio) was that SACD and DVD-Audio were both such catastrophic failures that only audiophiles even know what they were. Because in the mean time, mp3s took over.
And if VHS was 'technically inferior' to Betamax, mp3 compared to SACD/DVD-A would be off the charts bad. And guess what people chose?
You get Wii Sports with the console, and you get Wii Play when you buy a second controller. That in and of itself should be enough to explain how they outsell Zelda overall, especially since a lot of non-traditional gamer types are buying these things, many of whom have never heard of Zelda. I imagine very few at-launch buyers who do buy Zelda don't buy a second controller at the same time, so there's no way for Zelda to gain numbers on Wii Play and Wii Sports.
If Zelda doesn't sell a lot more than any non-bundled launch title, though, I will be very surprised.
> As for the '30 game rate to break even' 30 games isn't that many
> (Should imagine they're are quite a number of people with 30+ ps2
> games), and don't forget Sony get money on Blu-ray discs too.
I don't know... "quite a number" I will grant you. But more than half? More importantly, all things considered, does the mean work out to 30 or more? I would seriously doubt that.
You have a point on Blu-ray, but it may be a catch-22: Sony intended for the PS3 to rocket Blu-ray to the top as the new standard. If the PS3 fails, though, Blu-ray may not end up meaning much more than UMD.
That's right--before that if you were a subscriber you would've seen:
Posted by Hemos in the Mysterious Future
Subscribers can see the stories before the riff-raff (but posting is only possible after the official publication). This actually works out well for everyone, since it increases the chances that some of the early comments might be actually be on-topic and thought through rather than endless 'first post' mispellings. And subscribers can visit sites before they're slashdotted.
I meant to mod you 'insightful' (since my wife behaves the same way with those two games) but my finger slipped and it hit 'redundant' instead. There is no way to undo a moderation directly, but by posting in this discussion, my moderation will be undone and all will be set right. Please disregard this comment, everybody!
In fairness to GP, any analogy breaks down if you push it far enough, and proxy wars are a well-known phenomenon, of which the Korean war was one example. They don't have to be between superpowers either: witness the US-armed and Iranian-armed conflict we just had in South Lebanon...
Sonic Rush is a great 'classical' 2D platformer, and makes incredible use of the dual screens. Its graphics are good enough that it's one of my 'show off the system' titles, and the game is fun whether you have 2 minutes or half an hour to play.
Just FYI (since your post is totally clear given the context): "Shinobi" isn't the name of the character in the Shinobi series of games--it's a common Japanese word which is a synonym for "ninja". So the correct term would be "the ninja from the Shinobi games"--just like "Contra" isn't the name of the character either, so you'd have to say "the soldier from the Contra games" (which incidently, I would add to your list, along with Alex Kid).
I haven't tried it but have heard good things about WebHuddle, which is actually aimed for business meetings (you can even show slides, although for a whiteboard just leave a blank slide and scribble on that). Being aimed at a non-geek audience, setup is intended to be a no-brainer.
You need to know *exactly* what your server needs to serve.
A Mac Mini sure seems like the coolest possible "suitcase server" imaginable, but if you want or might want commercial "enterprise" products like DB2, Oracle, Sybase, WebLogic or WebSphere, your list of options gets quickly shortened for you. Neither OS X or Linux PPC are going to do the job, because you rely on closed-source software whose 'Linux support' is implicitly x86 only (plus *maybe* x86-64). Wait for the Intel-based mini I suppose, and make sure first that the packages you need will run on it.
But as I imagine others will point out, a 1U server for development work would be huge overkill; a remote server or simply running everything on your development machine (through VMWare if necessary) makes way, WAY more sense here, for a myriad of reasons. On the other hand, if your "server" world is 100% open source, you may well be able to craft the ultimate travel server exactly to your specifications with a geek-cool factor that is off the scale. But you don't *need* to; I've written AS/400 WebSphere apps with only Tomcat on Windows (sadly)... but hey, if that's the kind of work you're doing, good luck fitting an AS/400 into your briefcase...
Very insightful post. Obvious when you think about it, but the kind of thing that doesn't occur to you to think through when you're looking at the box in the store. But that has got to be a big factor in the let-down of multiplayer games in modern settings.
Look at the interactive fiction community--lots of great stuff is getting put out for free, and there's a non-trivial amount of coding involved in putting out an IF game.
But getting from that great story/gameplay to a 3d graphics-and-sound fest that would draw in casual gamers is a *huge* amount of work that would require many developer hours, and very little in the way of artistic creativity.
I think that this parallels a wider trend one sees in Open Source where interesting projects tend to be extremely succesful and dull/monotonous areas tend not to be addressed so well. Creativity is not lacking in open source games, as the IF community shows, but marshalling the resources necessary to make compelling 3d graphical games is beyond the community, due to the monotonous nature of the work.
This has indeed been around for years, but it is an opportunity to point out an advantage of the program which I have benefitted from that may be of interest to other slashdotters: yes, it is an opportunity to get to know VMS, HP/UX, and Tru64 Unix (and a few Linux distros, but you have no excuse for not running those on your own hardware). Granted, given the current shift in direction of HP away from (non-Windows based) server tech it is probably not that valuable as career knowledge to familiarise yourself with these OSes, but is still neat for people interested in exploring different operating systems.
But the Test Drive program also gives you access to compilers for a few languages that are tricky to find or unavailible in free software. IIRC you get Fortran 95 on Unix (which in 2006 may be less of a big deal than it was awhile back when there were no decent Free compilers for fortran, as I understand there has been progress on that front lately in gcc-land), and on VMS (if it's still the case--I whined specifically for this back when Compaq was running the program and they added it) you get a COBOL compiler--which there is still no full-featured Free software version of (yes, probably for good reason). Nonetheless, people interested in learning Fortran 95 or Cobol may find the TestDrive program to be the most accessible way of gaining access to these languages.
(And a third benefit, IIRC a few of the servers have Oracle already set up for you, which may be handy for people wanting to learn Oracle but unwilling/unsuccesful in installing it themselves. Although the argument could be made that managing to install Oracle yourself should be the first step in going about learning it...)
While the margins IBM makes on their i5 and z9 servers are huge, they actually do do a lot of dragging into the future as well. Both architectures are sold under two pricing schemes, one with all the legacy support and green screens enabled, and the other with only modern protocols enabled.
Companies that need the legacy support pay roughly ten times what you pay if you only use the modern (Java, WebSphere and DB2) stuff. So although they provide it, and profit heavily from it, they still also heavily encourage switching to Java and WebSphere rather than COBOL and CICS. They just allow companies to move once the price gap gets to the point that it becomes in their economic interest to do so instead of forcing them. "Dragging your customers into the future" has to be kept very distinct from "chasing them away". But they are turning up the heat more and more through licensing.
You can use Java with GTK, Cocoa, or native Windows toolkits. The question then becomes, why Java programmers are not interested.
The basic reason for this is, your application's GUI becomes completely unportable, and you suddenly have little reason for having written it in Java in the first place instead of the platform's "standard" language (C, Objective C, or C#).
A "middle way" is to do what SWT does and wrap the native widgets with a generic API. This creates a "lowest common denominator" problem, however, since you inevitably have to stick to only using those widgets which all your target platforms have...
The reason we see famine in Africa is that Africa has increased its population by 10 in the last 70-80 years.
Really? 10 isn't that many people for Africa to have increased its population by. Heck, I know of individual families that have ten kids in them, and you never hear of it causing a crisis for their town, let alone a whole continent.
Yikes.
As you evidently did not recognise my nick, it is Russian, which I speak fluently and (in addition, since speaking a language doesn't give one the knowledge of its origins) also know that it is descended from Slavonic, a language I also read, which in turn is also NOT descended from Greek (any more than Germanic, Persian, Italic, Sanskrit, or any other branches of our Indo-European tree). I've also spent an absurd number of years studying classics and am quite literate in Latin and Ancient Greek, and am a language geek generally (if I even needed to point that out at this point).
The tsar/caesar connexion is unrelated to linguistics, as are the reasons why Byzantine Greeks called themselves Romans. Just like Finland is often culturally grouped with the Scandinavian countries even though lingistically Swedish is more closely related to Urdu than Finnish. The kings of England claimed the throne of France for centuries too, and yet neither that nor the centuries of Norman rule before that could erase English's Germanic character and turn it into a Romance language.
You are correct that Byzantine civilisation had a lot to do with cementing the cultural influence of Greek, but the Indians were already speaking an Indo-European language for centuries before Alexander showed up there, as were the inhabitants of the European territories of the empire.
You are also correct that a huge amount of the scientific and medical terminology that people often call "Latin roots" are actually Greek roots, but such terms are neologisms anyway.
For fun I (unscientifically and hastily) counted the etymologies up in the last paragraph and got the following numbers (not counting duplicate words, and splitting "terminology" in half!):
Germanic: 14
Latin: 9.5
Greek: 1.5
* Only 'neologism' was fully Greek, ironically. Although the very next paragraph brought out "etymology" and "paragraph", but even so, not that many English words have Greek origins.
I'm sure wikipedia is a great resource to read up on a lot of this stuff, they have an article on "Historical linguistics".
Well, just to back up what your first corrector was saying in a little less flamey way, you are really, really wrong about that.
1. Most of English is not German. English is a Germanic language (as is German) but this ancestry does not account for most of our vocabulary. I read somewhere that only about 500 words of modern English can actually be traced back to Anglo-Saxon (though they are important words, and Anglo-Saxon had a much bigger influence on English grammar than the other languages we get our vocabulary from.
Analogy: My last name in English, but my blood is only 1/8 English since most of my (female) ancestors came from elsewhere. English is a Germanic language in the same sense.
2. Most of English vocabulary (>50%) is LATIN in origin, and most (but not all) of those words came into the language through FRENCH, which is itself descended from Latin.
3. Not only is German not a descendent of Greek, but neither is French, nor Latin, nor in fact ANY language other than modern Greek. This is what makes your assertion so funny to someone who has studied linguistics, because Greek is a very isolated branch of the Indo-European language family (although Greek culture being the foundation of Western Civilisation, there are many vocabulary words in all Western languages that are Greek borrowings. But no languages other than modern Greek are descended from Ancient Greek.)
An actual true statement is that English is a Germanic language and most of its vocabulary comes from Latin (through Norman French). (Although "most" in the previous sentence must mean in terms of number of words, not number of words weighted by frequency of usage.)
I can understand how someone might mix up German and Germanic (different languages, different eras, the similarity of names notwithstanding) and Latin and Greek (as pillars of our classical tradition), but the end result is spectacularly wrong, on so many levels.
I recommend reading up on linguistic history, it's a fascinating field and teaches you a lot about our world, our history, and our species.
1. OAS is not the leading product in the J2EE market either, and I would guess it's actually less popular than JBoss.
2. Total agreement about the oversaturation. IBM seems to be the dominant player all the same, however, and look what they're just starting to do after having bought Gluecode (a company that was putting out an easy-to-use install of the open-source Apache Geronimo): release it as "Websphere Community Edition", boosting their brand awareness at little cost to the company, and gearing up to provide well-supported migration to the 'big' WebSphere.
That strategy seems like a good one for building up market share in a saturated market, and Oracle might just have decided to copy it.
Heck, they already managed to change from driving on the left to driving on the right--that's more than most countries could pull off!
Obviously anyone who prefers modern Gnome or sees the reasoning behind it evolving the way it has and agrees with it should use Gnome. Someone who liked Gnome pre-2.2, when it had a much geekier feel (in terms of using Sawfish, being hackable in so many programming languages, etc.), faces a choice at some point. Either militate and try to gather support for pushing Gnome back in the direction it came from (an unattractive option since major companies like Novell and Sun will be pushing back, and it would seem to involve advocating moving the project backwards rather than forward), or else jump ship and switch to KDE or a window manager that suits his taste.
I would advocate the second option, both to save the Gnome project from unproductive whiners and allow it to make progress that is sure to be very beneficial to business deployments of GNU/Linux desktops, and to allow this hypothetical "ordinary geek" to be happier elsewhere.