For most competent folk, who spend alot of time on their computer, the mental adjustment period will be about an hour of use, in my experience.
And for a very large percentage (maybe 80% or so for me) of features, they will be immediately visible and obvious.
I think many people will find, like I did, that this is a much more intuitively organized interface, and just makes a lot more sense.
Yes, it will eliminate some of your burned-in muscle-memory for where certain functions are, but within a few uses, you will probably see net time saved because the interface is so much less deep, and just flat requires fewer clicks and movement than before.
And OH LORD the auto-formatting as you hover over choices. Remember in the old versions, how you wanted to test changes for a feature that is buried like 8 clicks deep. And how sometimes it would take 4 or 5 tries to get it looking how you like? This is just gone in the new interface, as the formatting temporarily applies as you hover over the formatting choice.
Give it a try, I think for the vast majority of people, within a month, they will be praising MS for the change. Seriously. Say what you will about Vista, the ribbon interface is something they got right, hit it right on the head.
No argument there. I must have 100,000 tiny.java files just in my c:\java\ hierarchy alone. I'm sure my drive is highly fragmented. (Of course, its arguable that development environments may not add much to fragmentation because most source code files are smaller than a single block size on modern drives.)
The question is: who cares?
Fragmentation != bad (for typical cases)
In other words, a high level of fragmentation does not necessarily indicate a slower system.
Fragmentation is inevitable in a file system, and not really a problem except under certain very specific use-cases like I mentioned above. For most users (streaming media being a counterexample), you just flat wont see a difference in perceived system performance from a highly fragmented drive.
Most modern OS's use a targeted optimization to put the most commonly read files in a group at the fastest point and defragments there (windows calls it superfetch or something similar in XP). For the most part, these are system files accessed during boot. The rest of the time, a fragmented drive just wont be that noticeable.
For many non-technical folk, defragmenting the hard drive is basically the equivalent of tech-voodoo, its like waving a burning stick over your computer and chanting, and will have about as much chance of making a noticeable performance difference as defragmenting regularly.
At least thats been my experince doing corporate IT for ~10 years... at various times, we would see if we could measure performance differences after defragmenting, and it just wasnt there.
Now, that being said, I'm sure there are specific use-cases where it does make a difference. But in most cases, it doesnt. For example, in a busy IIS web server, turning off last-read timestamp updates on the file system can make a very noticeable performance difference, where defragmenting makes no noticeable difference.
A big part of this is done in Office 2007 (at least in the ribbon-enabled products). Just click or highlight something and hover your mouse for a half-second.
Up pops a pseudo-context window with the most commonly used commands right there hanging off your mouse, so you dont have to move all the way up to the ribbon.
It is, in my opinion, a reasonable middle ground between everything in the menus/ribbon and everything in the context menu, neither of which is ideal.
And if you really want what you're talking about, you can get closer to it in office 2007 by using the 'minimize the ribbon' which auto-hides the ribbon until you move your mouse up there and click on the tab headings.
For what it's worth, this is exactly how MS Office works.
When you insert a table into a word document, you get a miniature Excel spreadsheet embedded into your word document, and vice versa. When you insert a picture and double-click on it, you get the picture editor.
You can in fact draw text boxes, tables, pictures, word-art, etc etc into most office documents exactly as you describe, and you get the behavior and power of their parent app when you double-click on them.
Now I can't say whether this was there in Office95... as my memory isnt that good, but its definitely been that way for a while. Supporting that exact use-case was one of the primary drivers behind OLE.
Also, its important to note that WinFS will sit on top of NTFS, not replace it. NTFS (some minor version upgrade) will still be the underlying file system.
And to be honest, the only major thing missing from NTFS was transactional commits, and this is coming soon along with transactional registry reads/writes (Vista? can't remember). You got most of the benefits you care about from single-file/single-unit transactions with the journalling, which has always been there.
One that doesn't require manual defragmenting the hard drive (everybody else can do it...)
Why does this keep coming up with people?
NTFS doesnt really require defragmentation under typical scenarios. Yes, it will get fragmented, but it basically asymptotically approaches a finite fragmentation point and hovers around there.
'Regular Defrag' or 'Automatic Defrag' isnt necessary in a typical business scenario. In a corporate environment you dont defrag your servers, nor do you your desktops. It's just really not necessary except under certain circumstances where you rely on high performance sequential/streaming reads.
Grrr, hit the submit button too early. To continue:
NT5 was not the first bottom-up re-write on NT... it was a release after 7 years of evolution from the NT team. And it wasnt based on ME, it was Windows 2000.
ME was the last dying gasp of the consumer-focused Win9x line.
Also note that the first public release of NT as v3.1 was a marketing choice to appear like it was evolving in line with the 'Windows' line, but there was no connection whatsoever at that point. It was basically a couple years of work from the VMS team that MS poached to create NT.
Vista is the culmination of ~10 years of evolution of the NT line, combined with some technologies brought in from the 9x line in the NT5/win2000 days. The VMS-like base that NT started with was solid, but has arguably been polluted by various strategic choices MS has made (marketing, consumer focus, backwards compatibility, GUI focus, etc).
You're randomly mixing the Win9x line with the NT line. Plus you've completely skipped 5.2 (ie, Windows 2003 Server).
From concise windows timeline, it looks more like this:
1985 - DOS based - Windows 1.0
1987 - DOS based - Windows 2.0
1990 - DOS based - Windows 3.0
1992 - DOS based - Windows For Workgroups 3.1
1993 - NT Kernel - NT 3.1
1994 - NT Kernel - NT 3.5
1995 - NT Kernel - NT 3.51
1995 - Win9x Core - Windows 95
1996 - NT Kernel - NT 4.0
1998 - Win9x Core - Windows 98
1999 - Win9x Core - Windows 98 SE
2000 - Win9x Core - Windows ME (last of the 9x based systems)
2000 - NT Kernel - NT 5.0 (ie, Windows 2000)
2001 - NT Kernel - NT 5.1 (ie, Windows XP)
2003 - NT Kernel - NT 5.2 (ie, Windows 2003 server)
2004 - NT Kernel - NT 5.1R2 (ie, Windows XP SP2)
2005 - NT Kernel - NT 5.2R2 (ie, Windows 2003 Server R2)
Thanks for the informative reply Pinky. As it happens, a while after the last post, I ran across the wikipedia entry, which was unexpectedly informative, and learned much of this.
And you're right, nearly all of the pain I've experienced with JavaScript has been DOM stuff, not core language features. Was a good read, and I appreciate the response.
Thanks!
PS, you're right about the offtopic, the parent just really tickled an irritation-bone in me, so felt the need to respond. But it was a bit offtopic.
I've got to second this, and would mod it up furiously if I hadnt already posted to this topic.
Where is the reference to javascript? Who owns it? When I google for javascript reference, the best thing I get is mozilla's site... but they dont make it clear whether their reference is to their implementation of javascript, or to the language spec, or to the actual reference implementation (assuming there even exists a reference implementation).
And no one seems to actually own it and be the master of it. The closest I can find is ECMAScript, but ECMAScript isnt JavaScript, rather JavaScript is based on ECMAScript (ie, a superset of).
I would love to hear an authoritative answer to this, but in all of my professional career, have never found one.
I bet 10 bucks that Mr. Eckel's 3D card drivers are out of date or not installed. The application he linked to uses JOGL (Java OpenGL bindings), so if his computer is unable to run OpenGL, he will be unable to run the app. It's a rare issue, but it happens. The easy way of debugging the app is to either bring up the Java Console through the Coffee Cup in the system tray, or to go into the Java Web Start settings and enable the console (or logging!) there. Easy, peasy.:)
For what it's worth, what he describes is exactly what happens to my machine. It spends a few minutes downloading stuff furiously, and then just... nothing. No feedback, no error messages, not a single thing.
I'm not sure who's fault it is in this case, whether Sun's or the folks at the website, but this should never happen. If there is a dependency on some level of OpenGL support, then the software or platform should test that, and complain to the user (with instructions on how to fix it) before running.
To just fail silently, with no feedback whatsoever, is never, ever acceptable.
At least 40% of the population can be trained to do any job. 110 IQ is plenty for just about any job on the planet.
Where do you get this information? How exactly was this experimented on?
Because frankly, it looks like you just pulled some random numbers out of your butt and used them as truth to support your arguments.
And I can definitely refute this idea that 'training' can turn an unqualified person into a qualified person. I've worked with employees I've inherited in past jobs that while they had all the best intentions, and took advantage of all the training they could get, they just flat couldnt keep up. They didnt have the right kind of brain to work in software development.
And there are brilliant people who can't hold down jobs, because their attitude and productivity is so terrible, they're worse than 'average' people. You cant teach a good attitude, or strong work ethics.
Thats not to say that there aren't classes or training for these kinds of characteristics, its just that sending someone to them who isnt interested in learning is pointless. Nothing will change.
And for those folks who want to have good attitudes, and who would voluntarily go to these sorts of things, you dont really need training. Because lets be honest, having that desire to have a good attitude is most of that good attitude.
Much like Marxism... while every human has infinite theoretical potential, and could theoretically do decent at anything... it doesnt work that way in reality. Not everyone wants to learn, or cares enough to try, or is willing to open their mind up enough to grow.
I disagree completely. There are plenty of people out there who are just not born to be programmers.
They lack the ability to 'smell bad code' and improve it. They lack the ability to hold large models and relationships in their heads at once.
These are, in many ways, akin to how some people have a very natural ability to 'get' math and logic, and some don't.
In addition, even for those skills that are trainiable... the person has to want to learn, and take an active and interested role in the learning. You cant just throw 'training' at someone and make them competent.
In other words: There are things that can be learned, but cannot be taught.
Because a crappy programmer with lots of training is still a crappy programmer, he just knows lots of factoids.
Training on specific skills is the easiest thing to fix. What is hard is to find quality people, who will invest in the work, take pride in what they do, be able to communicate and collaborate with others, have a positive attitude, etc.
For the most part, those arent things that can be taught.
Except thats not what happens. Within about 5 seconds, your computer shifts into offline mode for that server, and uses the locally cached copy of documents, and re-synchs next time you are connected.
The only reason why this wouldnt happen by default is if Offline File was explicitly turned off to prevent this. There are scenarios where you'd want that, but your Redirected MyDocuments isnt usually one of them.
Exchange is not that complicated to setup and maintain. There is a learning curve because its a large server product, but its not that bad.
And if you're a small business getting it in SBS form, then its literally trivial.
In my experience the biggest driver for outsourcing is not to reduce the complexity, its to make the cost reasonable for the ultra-mission-critical product that email server is.
For many businesses, their email and calendaring is so critically important, that no downtime is allowable. But for a small business, to pay for the staff and hardware to setup a proper clustered front end and a highly fault tolerant shared disk environment, plus the regular test/prove of backup/restore is not within the realm of reality.
There are plenty of consultants with the time and skills to do it, but the cost is high. Compare that to $15 per month per user for a quality, private-branded exchange account like intermedia.net, and there's no comparison.
What does Exchange have to do with replacing Sun Workstations with Wintel workstations?
If you had Exchange before, then the same people that handled exchange before should have continued to handle Exchange.
If you didnt have Exchange before, then we're not talking about replacing one kind of workstation with another, but a rip n' replace of the entire back-end infrastructure, and by incompetent people from the sounds of it.
... Windows PC's that were always slowing up or crashing, especially after that stupid automated windows update.
Why were the windows machines 'slowing up' or crashing? If the patches caused slowdowns or crashes, then why did your windows admins release the patches? Did they then roll back the patches?
Not trying to be difficult here, but it sounds like this was much more than just replacing sun workstations with wintel workstations, it sounds like a complete change of infrastructure, with some not very good new windows admins.
Was it a change to an outsource company, by any chance? CSC maybe?
There's a bunch of them. Internet2/Abilene, Lambda Rail, CENIC, and a ton of point-to-point high speed connections between universities, or universities and labs, etc.
Many big corporations also have their own cross-country private 'internets'.
So I understand the DirectX 10 reference, but have no clue what you're talking about with Office 2007.
I'm running Office 2007 (RTM, not a beta) right now on my XP Pro box, and it works like a charm. In fact, based on what I've seen so far... Office 2007 is something they genuinely did a good job at. The ribbon interface is quite nice, and a massive improvement over the deeply nested ones before.
In fact, as best I've seen, the only difference between Office 2007 on XP and Office 2007 on Vista is something called 'instant search' in outlook 2007 that supposedly works much better on Vista.
Vista is stealing the next generation of hardware from us.
How do you figure this? The very paragraph you quote right above that says the opposite. It basically says that MS is forced into a faustian bargain by the content owners; either: 1. Put this stuff in and give Vista users the ability to play blue-ray, hd-dvd, etc, or 2. Dont put this stuff in which will prevent Vista users from accessing this content. And I dont blame them for making this choice. Regular Vista users (ie, non-techie folk) wont understand or care why they cant play the discs, they'll just think its MS fault. And so MS takes the flak from the content owners and distributers.
... an odf/xml feature would be trivial to add, but MS flatly refused to make a plugin for Office to convert to odf/xml, even though it meant losing the state's patronage.
Microsoft is really determined to strangle open formats.
Yes, so determined that they're actually paying for someone else to do it and host it on sourceforge.
Slashdot Mirror
I've got to second ocbwilg's comments.
For most competent folk, who spend alot of time on their computer, the mental adjustment period will be about an hour of use, in my experience.
And for a very large percentage (maybe 80% or so for me) of features, they will be immediately visible and obvious.
I think many people will find, like I did, that this is a much more intuitively organized interface, and just makes a lot more sense.
Yes, it will eliminate some of your burned-in muscle-memory for where certain functions are, but within a few uses, you will probably see net time saved because the interface is so much less deep, and just flat requires fewer clicks and movement than before.
And OH LORD the auto-formatting as you hover over choices. Remember in the old versions, how you wanted to test changes for a feature that is buried like 8 clicks deep. And how sometimes it would take 4 or 5 tries to get it looking how you like? This is just gone in the new interface, as the formatting temporarily applies as you hover over the formatting choice.
Give it a try, I think for the vast majority of people, within a month, they will be praising MS for the change. Seriously. Say what you will about Vista, the ribbon interface is something they got right, hit it right on the head.
No argument there. I must have 100,000 tiny .java files just in my c:\java\ hierarchy alone. I'm sure my drive is highly fragmented. (Of course, its arguable that development environments may not add much to fragmentation because most source code files are smaller than a single block size on modern drives.)
... at various times, we would see if we could measure performance differences after defragmenting, and it just wasnt there.
The question is: who cares?
Fragmentation != bad (for typical cases)
In other words, a high level of fragmentation does not necessarily indicate a slower system.
Fragmentation is inevitable in a file system, and not really a problem except under certain very specific use-cases like I mentioned above. For most users (streaming media being a counterexample), you just flat wont see a difference in perceived system performance from a highly fragmented drive.
Most modern OS's use a targeted optimization to put the most commonly read files in a group at the fastest point and defragments there (windows calls it superfetch or something similar in XP). For the most part, these are system files accessed during boot. The rest of the time, a fragmented drive just wont be that noticeable.
For many non-technical folk, defragmenting the hard drive is basically the equivalent of tech-voodoo, its like waving a burning stick over your computer and chanting, and will have about as much chance of making a noticeable performance difference as defragmenting regularly.
At least thats been my experince doing corporate IT for ~10 years
Now, that being said, I'm sure there are specific use-cases where it does make a difference. But in most cases, it doesnt. For example, in a busy IIS web server, turning off last-read timestamp updates on the file system can make a very noticeable performance difference, where defragmenting makes no noticeable difference.
A big part of this is done in Office 2007 (at least in the ribbon-enabled products). Just click or highlight something and hover your mouse for a half-second.
Up pops a pseudo-context window with the most commonly used commands right there hanging off your mouse, so you dont have to move all the way up to the ribbon.
It is, in my opinion, a reasonable middle ground between everything in the menus/ribbon and everything in the context menu, neither of which is ideal.
And if you really want what you're talking about, you can get closer to it in office 2007 by using the 'minimize the ribbon' which auto-hides the ribbon until you move your mouse up there and click on the tab headings.
For what it's worth, this is exactly how MS Office works.
... as my memory isnt that good, but its definitely been that way for a while. Supporting that exact use-case was one of the primary drivers behind OLE.
When you insert a table into a word document, you get a miniature Excel spreadsheet embedded into your word document, and vice versa. When you insert a picture and double-click on it, you get the picture editor.
You can in fact draw text boxes, tables, pictures, word-art, etc etc into most office documents exactly as you describe, and you get the behavior and power of their parent app when you double-click on them.
Now I can't say whether this was there in Office95
Also, its important to note that WinFS will sit on top of NTFS, not replace it. NTFS (some minor version upgrade) will still be the underlying file system.
And to be honest, the only major thing missing from NTFS was transactional commits, and this is coming soon along with transactional registry reads/writes (Vista? can't remember). You got most of the benefits you care about from single-file/single-unit transactions with the journalling, which has always been there.
Why does this keep coming up with people?
NTFS doesnt really require defragmentation under typical scenarios. Yes, it will get fragmented, but it basically asymptotically approaches a finite fragmentation point and hovers around there.
'Regular Defrag' or 'Automatic Defrag' isnt necessary in a typical business scenario. In a corporate environment you dont defrag your servers, nor do you your desktops. It's just really not necessary except under certain circumstances where you rely on high performance sequential/streaming reads.
Grrr, hit the submit button too early. To continue:
... it was a release after 7 years of evolution from the NT team. And it wasnt based on ME, it was Windows 2000.
NT5 was not the first bottom-up re-write on NT
ME was the last dying gasp of the consumer-focused Win9x line.
Also note that the first public release of NT as v3.1 was a marketing choice to appear like it was evolving in line with the 'Windows' line, but there was no connection whatsoever at that point. It was basically a couple years of work from the VMS team that MS poached to create NT.
Vista is the culmination of ~10 years of evolution of the NT line, combined with some technologies brought in from the 9x line in the NT5/win2000 days. The VMS-like base that NT started with was solid, but has arguably been polluted by various strategic choices MS has made (marketing, consumer focus, backwards compatibility, GUI focus, etc).
Your timeline is way off.
You're randomly mixing the Win9x line with the NT line. Plus you've completely skipped 5.2 (ie, Windows 2003 Server). From concise windows timeline, it looks more like this:
Thanks for the informative reply Pinky. As it happens, a while after the last post, I ran across the wikipedia entry, which was unexpectedly informative, and learned much of this.
And you're right, nearly all of the pain I've experienced with JavaScript has been DOM stuff, not core language features. Was a good read, and I appreciate the response.
Thanks!
PS, you're right about the offtopic, the parent just really tickled an irritation-bone in me, so felt the need to respond. But it was a bit offtopic.
I've got to second this, and would mod it up furiously if I hadnt already posted to this topic.
... but they dont make it clear whether their reference is to their implementation of javascript, or to the language spec, or to the actual reference implementation (assuming there even exists a reference implementation).
Where is the reference to javascript? Who owns it? When I google for javascript reference, the best thing I get is mozilla's site
And no one seems to actually own it and be the master of it. The closest I can find is ECMAScript, but ECMAScript isnt JavaScript, rather JavaScript is based on ECMAScript (ie, a superset of).
I would love to hear an authoritative answer to this, but in all of my professional career, have never found one.
For what it's worth, what he describes is exactly what happens to my machine. It spends a few minutes downloading stuff furiously, and then just ... nothing. No feedback, no error messages, not a single thing.
I'm not sure who's fault it is in this case, whether Sun's or the folks at the website, but this should never happen. If there is a dependency on some level of OpenGL support, then the software or platform should test that, and complain to the user (with instructions on how to fix it) before running.
To just fail silently, with no feedback whatsoever, is never, ever acceptable.
Where do you get this information? How exactly was this experimented on?
Because frankly, it looks like you just pulled some random numbers out of your butt and used them as truth to support your arguments.
And I can definitely refute this idea that 'training' can turn an unqualified person into a qualified person. I've worked with employees I've inherited in past jobs that while they had all the best intentions, and took advantage of all the training they could get, they just flat couldnt keep up. They didnt have the right kind of brain to work in software development.
And there are brilliant people who can't hold down jobs, because their attitude and productivity is so terrible, they're worse than 'average' people. You cant teach a good attitude, or strong work ethics.
Thats not to say that there aren't classes or training for these kinds of characteristics, its just that sending someone to them who isnt interested in learning is pointless. Nothing will change.
And for those folks who want to have good attitudes, and who would voluntarily go to these sorts of things, you dont really need training. Because lets be honest, having that desire to have a good attitude is most of that good attitude.
Much like Marxism ... while every human has infinite theoretical potential, and could theoretically do decent at anything ... it doesnt work that way in reality. Not everyone wants to learn, or cares enough to try, or is willing to open their mind up enough to grow.
I disagree completely. There are plenty of people out there who are just not born to be programmers.
... the person has to want to learn, and take an active and interested role in the learning. You cant just throw 'training' at someone and make them competent.
They lack the ability to 'smell bad code' and improve it. They lack the ability to hold large models and relationships in their heads at once.
These are, in many ways, akin to how some people have a very natural ability to 'get' math and logic, and some don't.
In addition, even for those skills that are trainiable
In other words: There are things that can be learned, but cannot be taught.
Because a crappy programmer with lots of training is still a crappy programmer, he just knows lots of factoids.
Training on specific skills is the easiest thing to fix. What is hard is to find quality people, who will invest in the work, take pride in what they do, be able to communicate and collaborate with others, have a positive attitude, etc.
For the most part, those arent things that can be taught.
Except thats not what happens. Within about 5 seconds, your computer shifts into offline mode for that server, and uses the locally cached copy of documents, and re-synchs next time you are connected.
The only reason why this wouldnt happen by default is if Offline File was explicitly turned off to prevent this. There are scenarios where you'd want that, but your Redirected MyDocuments isnt usually one of them.
You dont have MS Office anymore on your machine. So while you're offline, you're screwed.
Exchange is not that complicated to setup and maintain. There is a learning curve because its a large server product, but its not that bad.
And if you're a small business getting it in SBS form, then its literally trivial.
In my experience the biggest driver for outsourcing is not to reduce the complexity, its to make the cost reasonable for the ultra-mission-critical product that email server is.
For many businesses, their email and calendaring is so critically important, that no downtime is allowable. But for a small business, to pay for the staff and hardware to setup a proper clustered front end and a highly fault tolerant shared disk environment, plus the regular test/prove of backup/restore is not within the realm of reality.
There are plenty of consultants with the time and skills to do it, but the cost is high. Compare that to $15 per month per user for a quality, private-branded exchange account like intermedia.net, and there's no comparison.
Wait ... you mean the tagging feature is supposed to be useful?
I thought it was put in for humourous effect.
But then again, I also miss 'itsatrap'.
... needed to second-guess MS exchangeWhat does Exchange have to do with replacing Sun Workstations with Wintel workstations?
If you had Exchange before, then the same people that handled exchange before should have continued to handle Exchange.
If you didnt have Exchange before, then we're not talking about replacing one kind of workstation with another, but a rip n' replace of the entire back-end infrastructure, and by incompetent people from the sounds of it.
... Windows PC's that were always slowing up or crashing, especially after that stupid automated windows update.Why were the windows machines 'slowing up' or crashing? If the patches caused slowdowns or crashes, then why did your windows admins release the patches? Did they then roll back the patches?
Not trying to be difficult here, but it sounds like this was much more than just replacing sun workstations with wintel workstations, it sounds like a complete change of infrastructure, with some not very good new windows admins.
Was it a change to an outsource company, by any chance? CSC maybe?
There's a bunch of them. Internet2/Abilene, Lambda Rail, CENIC, and a ton of point-to-point high speed connections between universities, or universities and labs, etc.
Many big corporations also have their own cross-country private 'internets'.
So I understand the DirectX 10 reference, but have no clue what you're talking about with Office 2007.
... Office 2007 is something they genuinely did a good job at. The ribbon interface is quite nice, and a massive improvement over the deeply nested ones before.
I'm running Office 2007 (RTM, not a beta) right now on my XP Pro box, and it works like a charm. In fact, based on what I've seen so far
In fact, as best I've seen, the only difference between Office 2007 on XP and Office 2007 on Vista is something called 'instant search' in outlook 2007 that supposedly works much better on Vista.
How do you figure this? The very paragraph you quote right above that says the opposite. It basically says that MS is forced into a faustian bargain by the content owners; either: 1. Put this stuff in and give Vista users the ability to play blue-ray, hd-dvd, etc, or 2. Dont put this stuff in which will prevent Vista users from accessing this content. And I dont blame them for making this choice. Regular Vista users (ie, non-techie folk) wont understand or care why they cant play the discs, they'll just think its MS fault. And so MS takes the flak from the content owners and distributers.
... an odf/xml feature would be trivial to add, but MS flatly refused to make a plugin for Office to convert to odf/xml, even though it meant losing the state's patronage. Microsoft is really determined to strangle open formats.Yes, so determined that they're actually paying for someone else to do it and host it on sourceforge.
http://odf-converter.sourceforge.net/
Then why is using Java okay?
Integrating with Subversion is about a 5 minute process:
t /svn_integration.html
I RA+Subversion+plugin
t /trackback.html
http://www.atlassian.com/software/jira/docs/lates
http://confluence.atlassian.com/display/JIRAEXT/J
I'm not sure which wiki you're referring to by 'the wiki', but it integrates quite nicely with Confluence:
http://www.atlassian.com/confluence/
It also has a generic trackback mechanism to integrate with other wiki's:
http://www.atlassian.com/software/jira/docs/lates