Your D70 will be fine - ACR works just great with this.
However, to answer your question about plugins... Nikons RAW converter is a steaming pile of poo when compared to the glory that is ACR.
I'm sure Nikon's converters in Nikon Capture are geat, but I don't use this. I use Adobe Photoshop, because this way, whether I'm using Canon, Nikon or Minolta, I can still use the same app to do my post work.
Nikon's PS plugin is crap. It allows you to set whitebalance and tweak a couple of things. ACR on the other hand is the cornerstone of my digital workflow, and since learning to use ACR properly, I've cut my post time per picture from around 10 minutes to around 2. That's a five-fold speed improvement. If getting the latest Nikon camera means I have to multiply my post-processing time by 5, I guess I won't be buying a Nikon.
Adobe's ACR does things differently to how Nikon Capture would. This includes colour choices, white balance, etc.
By forcing you to use their SDK or break the law, Nikon are saying that their's is the right (and only) way to decode NEF files.
By using the SDK, you are limited to the fastest it can process files - you can never process a file faster by working out your own solutions, you're always limited to the speed of the libraries.
You're also forced to live with all the bugs, memory leaks, etc. You never really have the kind of control that you want over your app.
If the job at hand was hard, then it might make sense for an API to help you, but everyone manages this work just fine. So why force an SDK on us ?
Their SDK is not useful to many people, specifically people who code for Linux. Besides licencing issues, etc. I'm not sure that the libraries will run on Linux.
In any event, it's not hugely relevant anymore - Dave Coffin has broken the encryption so we can all go on with our lives:)
I keep hearing how brilliant bitkeeper was because it made Linus more productive. From what I can tell, this could have been done without bitkeeper, but due to stubornness, it tool a technical tool.
Apoligies in advance if I get some of this wrong, but here is how I understand the issue...
Linus was the Keeper of the Keys - no code got to the kernel but through him. The problem was that he was dropping patches. So responsibility for various areas of the kernel was delegated out to luitenants, but they actually spelled it correctly;) These people were supposed to be the filter to make sure code was sane before it got to Linus.
This is where my timeline gets fuzzy... Either this delegated model happened at the same time as bitkeeper, or a bit earlier. What I do remember though is that Linus was checking every single patch before accepting it to the kernel. He didn't trust his 2ICs enough to just accept their code.
What bitkeeper did to make Linus more productive could have been done by Linus if he were more trusting. Bitkeeper took away his ability to just pick and choose individual patches, and forced him to accept entire changesets.
By doing this, bitkeeper forced Linux to trust these other people to have vetted the code. He could now either take the entire changeset or none of it.
This trust in these additional people stopped him doing the very time consuming job of checking each and every patch before accepting it, leaving him free to do other things.
So what bitkeeper did to make Linus more productive, Linus could have done without a software tool.
No, you're not. I'm actually holding off development of a new app because I'm waiting for this.
It will cut my development time by days and I'll be able to have a prototype out the door in a week. I really love the way it hooks into Interface Builder so that even during the dev stage, you can just lob a view to your data in there, and see how you think it will work best for the user.
The fact that you can save to something like sqllite means that it will be trivial to slap a web front end on your app as well.
I recently bought a Dyson vacuum cleaner and it had this little 'history' booklet attached to the unit. The last page of the booklet is titled 'The Patent Nightmare'
Apparantly people who make Real Things(tm) have problems with patents too!
But it still seems to me that a company that offers predictably not-bad food with low prices, quick service and convenient locations
Uhm, do you know the cost of that convinience ? Do you know the damage that McDonalds has done to food (not just food they serve, but all meat you buy)?
The fast food industry are the main reason that things like Ecoli is now so common in store bought foods. They are also responsible for huge labour issues, etc. They are the company the sue activists and then try to get the case canned because they are taking such a beating on the PR front.
Go read fast food nation. Personally, I think that Darfur has a better history!
I would say that Novell are definitely worth supporting, if only as a stick to beat Red Hat with.
Red Hat's continued virtual monopoly of the enterprise space has made them arrogant to a degree, and a little lazy. Having Novell around gives us a check against them.
Ok, Red Hat have their RHN solution. They also provide a tool named Satellite which effectively provides RHN on your own site. The idea is that it will manage your entire environment from deployment, through patching, rollback, configuration management, etc. On paper and in the demos it looks like an amazing product.
Having said that, it's looking increasingly likely that this pile of shit will cost me my job. From small problems like being able to take the entire Satellite solution down with a single security scan, to half implemented features, to poor support, I'm currently fighting a rear guard action trying to justify this product where I recommended it. You can't have multiple tiers with the overview information all being visible at the top of the tree like you can with SMS, etc.
I've learnt one very valuable lesson from all of this - People who laugh at people who say no-one ever got fired for buying Microsoft, obviously haven't been on the firing line for buying something else.
Ignoring Satellite, that pretty much leaves the following: 1. Their support period - they guarantee to support the distro for 7 years now
2. Guarantee of API / ABI stability - This is great on paper, but doesn't happen. It didn't happen in RHEL3 because of the changes to Linux Threads, but a simple fix with LD_ASSUME_KERNEL was available. Still, it scared managers. I have no idea how they plan to do this in RHEL4 because of the new kernel development process with so much change happening in the 'stable' kernel.
3. Support offering - so far, IMHO, this is shit. I've had such gems as being told that I shouldn't hot-swap disks in an HP-DL380, but that I should power the machine down and replace the disks. I'm so glad I have that in writing - makes for great stories at parties:)
4. SELinux in RHEL4 - This looks good, but I think it will be a while before enterprise apps such as Oracle RAC work seamlessly with this.
5. AD integration - They are providing out of the box AD integration with RHEL4 but it's pretty primitive. It just uses Winbind and this isn't suitable for a lot of bigger oranisations with more complex AD heirachys.
Overall, I'm not impressed with Linux in the Enterprise. Having dealt with Red Hat and Novell, even though I'm a dyed in the wool Linux geek, I still feel more confident about Solaris. Even with Sun's schizophrenia:)
I've written to every one of my MEP's and got responses from 2 of them.
The response from the Liberal Democrat was quite good. She sits on the JURI council who voted to restart patents, and isn't happy with the outcome. She was quite optimistic several weeks ago when I started corresponding with her, but this has changed.
The FastFood^WNew labour candidate is fully in support of them, so I'm casting a punishment vote in the upcoming general election.
What about countries where English isn't that common? In those places, MS will have to run local call centres and that will push their costs up. Surely this increase in costs will be more than the savings from thwarting privacy ?
Red Hat are probably the most closed company I've ever seen. They may still do a lot of development and contribute in code and investment (hardware certification, etc.) but at a personal level, I'd rather deal with Microsoft than with Red Hat.
When Windows server 2003 launched, I needed to get familiar with it, so I went to Microsoft's site and downloaded a 120 day eval copy. Using this, I was able to get familiar with the changes, etc.
When RHEL3 came out, I needed to do the same. Now, I'm an RHCE, so I've paid them a lot of my money directly (600 quid just for the exam), and sent even more money their way by recommending Red Hat wherever I went.
Do you think I could get an eval ? Fuck no! I called the Guildford office in London, I called the head office in the USA, I e-mailed, I tried everything.
Two months later, I had an interview at an investment bank, and they asked if I had experience with RHEL3. I explained that I didn't, but I had experience with RHEL 2.1AS as well as their free versions.
Microsoft 'get it' - they realise that if they help me learn their products, I can get a job using their products and they will sell more of their products. And I've never given MS a penny for training or certification.
Red Hat ? This bastion of freedom, this shining light of openness won't let me trial their products - they won't give you a thing to help you unless you pony up more money.
Sure, I could have spent the $200 on RHEL3 workstation, but with that attitude why should I? SuSE were quite happy to provide a trial version of SLES and now that's what I sell to all of my clients. I don't lose out, SuSE gains, and Red Hat lose out on a couple of enterprise contracts. It's all good:)
Practically nobody? You mean nobody who uses x86 technology. Sun Ultra Sparc 4 chips have been dual core for quite some time now. I believe that IBM also has dual cores in their power series.
Everything I've read so far says that two separate chips will give better performance than a dual core at the same clock speeds.
So if you have a dual Xeon 3.6Ghz, you're likely to get better performance than a machine with a single dual core 3.6Ghz.
This comes down to cores having to wait for access to resources, etc.
This is why I don't like the dual core == dual licence scheme. I'm _NOT_ getting twice the performance as with a single chip, but I have to pay twice.
In fact, this is something that makes Fujitsu servers attractive as competition for Sun. You can get equivalent performance to a dual core Sun Sparc IV 1.25Ghz with a single 1.8Ghz Fujitsu Sparc processor. Those clock speeds might be slightly out, but find the nearest:) So not only are you getting the processor cheaper, you're HALVING your licence costs.
Remember, it's not just a few players in the enterprise market that licence like this. Veritas, Oracle, HP Openview, Websphere MQ, they all do this. So if you can get the same performance from a single core CPU as you can from a dual core, halving your licence costs can be a big deal!
There are many cases of Microsoft deviating from accepted standards. The reasons that are normally given for this generally don't stand up to public scrutiny.
Why does Microsoft persist in breaking standards just to lock the competition out? Is it that you are too scared to compete on the merits of your products, the fact that this behaviour has become institutionalised, or some other reason?
Winbind is worthless for most real purposes. You cannot assign a UID to a user and there is no guarantee that a user will have the same uid each time they log on.
If I log on now, and get a UID of 1000, then at some point changes are made to the AD server, or a new version of samba is installed that supports computer accounts, when I next log-in, I will not get that same UID.
There are also issues with groups.
LDAP can be used quite effectively, but again there are limitations and issues. I have got around most of these, but not all of them.
Who said anything about the race of the professional car thieves ? If you scratch beneath the surface you'll find more than just a few of the syndicates are white owned and controlled.
So what does the lesson become then? That the Brits should have won the boer war and kept those uppity farmers in their place? Yeah - that sounds reasonable. It's not like a British subject was recently convicted in South Africa for participating in the planning of a coup. Oh, wait it is.
Maybe then the moral should be that you can't blame entire demographics for _anything_ and that you actually need to look at individuals.
In some countries, car theft is not just something that happens occasionally - it's an industry. And as in all industries, there are the rank amatuers and the pro's - For the pro's, this looks like a good option.
Consider South Africa - an entire arms race grew up around car theft. First the thieves just took cars when they were parked, so the insurance companies insisted that everyone have alarms and immobilisers.
The thieves got around those pretty quick - rumour is that a lot of professional's signed up for work at installation centres, learnt their way around them, and went back to work.
Next step was the gearlock - a device that locks the gearstick into a specific gear. IIRC, you couldn't remove the key on the earlier units unless you had the gearlock in, and if your car was stolen, the insurance company insisted on seeing all 3 keys.
Now with cars being so hard to steal, the age of the hi-jack was ushered in. If they can't get your car while it's parked, they'll take it while it's roll rolling.
In response, anti hi-jack systems became the norm. I can't remember how it was activated, but basically the bad guys show up, you let them take the car, they roll 20 metres down the road and the car cuts out and an alarm starts going off.
Around the same time we also go Satellite tracking, although I seem to remember something about it actually using the cellular infrastructure (GSM) not satellite - I may be wrong on this. Initially, the recovery rate on stolen and hi-jacked cars went through the roof. Unfortunately, the bad guys just upped the stakes. Soon we started seeing more kidnappings and murders as part of hi-jacks because if you can't call the stolen car in, they have longer to chop it.
Many vehicles were stolen to order, and not just new cars. Older cars that were common on the road were often targetted, then broken for spares. Cars that you wouldn't normally think twice about were stolen for export to Botswana and Zimbabwe, because the availability of spares for these made them popular vehicles.
Of my close circle of family and friends, we have had at least 10 cars stolen. Of those, not a single one has been recovered, so it's not a huge risk occupation really:)
I'm willing to bet that if this flaw is used anywhere, it will be used in South Africa - it's just one more tool for the biggest growth industry around:)
Sun says that customers are demanding it, but I wonder how real that demand is...
I know for a fact that Sun have sent letters to some customers asking them to ask ISVs about support for solaris 10. From what I can tell, this is to try and create the illusion of a grassroots movement where none exists.
So in light of this, how much of this 'demand' is actually real, and how much of it is Sun customers doing what Sun tell them in the hopes of a more favourable relationship when their support contract comes up for renewal ?
Of course, if you have an iPod, you could just buy iLingo which is a portable phrase book for the iPod. It gives many phrases and apparantly will speak them for you as well.
My wife bought an iPod purely because of this application!
Slashdot Mirror
Your D70 will be fine - ACR works just great with this.
However, to answer your question about plugins... Nikons RAW converter is a steaming pile of poo when compared to the glory that is ACR.
I'm sure Nikon's converters in Nikon Capture are geat, but I don't use this. I use Adobe Photoshop, because this way, whether I'm using Canon, Nikon or Minolta, I can still use the same app to do my post work.
Nikon's PS plugin is crap. It allows you to set whitebalance and tweak a couple of things. ACR on the other hand is the cornerstone of my digital workflow, and since learning to use ACR properly, I've cut my post time per picture from around 10 minutes to around 2. That's a five-fold speed improvement. If getting the latest Nikon camera means I have to multiply my post-processing time by 5, I guess I won't be buying a Nikon.
Adobe's ACR does things differently to how Nikon Capture would. This includes colour choices, white balance, etc.
By forcing you to use their SDK or break the law, Nikon are saying that their's is the right (and only) way to decode NEF files.
By using the SDK, you are limited to the fastest it can process files - you can never process a file faster by working out your own solutions, you're always limited to the speed of the libraries.
You're also forced to live with all the bugs, memory leaks, etc. You never really have the kind of control that you want over your app.
If the job at hand was hard, then it might make sense for an API to help you, but everyone manages this work just fine. So why force an SDK on us ?
Their SDK is not useful to many people, specifically people who code for Linux. Besides licencing issues, etc. I'm not sure that the libraries will run on Linux.
:)
In any event, it's not hugely relevant anymore - Dave Coffin has broken the encryption so we can all go on with our lives
I keep hearing how brilliant bitkeeper was because it made Linus more productive. From what I can tell, this could have been done without bitkeeper, but due to stubornness, it tool a technical tool.
;) These people were supposed to be the filter to make sure code was sane before it got to Linus.
Apoligies in advance if I get some of this wrong, but here is how I understand the issue...
Linus was the Keeper of the Keys - no code got to the kernel but through him. The problem was that he was dropping patches. So responsibility for various areas of the kernel was delegated out to luitenants, but they actually spelled it correctly
This is where my timeline gets fuzzy... Either this delegated model happened at the same time as bitkeeper, or a bit earlier. What I do remember though is that Linus was checking every single patch before accepting it to the kernel. He didn't trust his 2ICs enough to just accept their code.
What bitkeeper did to make Linus more productive could have been done by Linus if he were more trusting. Bitkeeper took away his ability to just pick and choose individual patches, and forced him to accept entire changesets.
By doing this, bitkeeper forced Linux to trust these other people to have vetted the code. He could now either take the entire changeset or none of it.
This trust in these additional people stopped him doing the very time consuming job of checking each and every patch before accepting it, leaving him free to do other things.
So what bitkeeper did to make Linus more productive, Linus could have done without a software tool.
No, you're not. I'm actually holding off development of a new app because I'm waiting for this.
It will cut my development time by days and I'll be able to have a prototype out the door in a week. I really love the way it hooks into Interface Builder so that even during the dev stage, you can just lob a view to your data in there, and see how you think it will work best for the user.
The fact that you can save to something like sqllite means that it will be trivial to slap a web front end on your app as well.
I recently bought a Dyson vacuum cleaner and it had this little 'history' booklet attached to the unit. The last page of the booklet is titled 'The Patent Nightmare'
Apparantly people who make Real Things(tm) have problems with patents too!
But it still seems to me that a company that offers predictably not-bad food with low prices, quick service and convenient locations
Uhm, do you know the cost of that convinience ? Do you know the damage that McDonalds has done to food (not just food they serve, but all meat you buy)?
The fast food industry are the main reason that things like Ecoli is now so common in store bought foods. They are also responsible for huge labour issues, etc. They are the company the sue activists and then try to get the case canned because they are taking such a beating on the PR front.
Go read fast food nation. Personally, I think that Darfur has a better history!
I would say that Novell are definitely worth supporting, if only as a stick to beat Red Hat with.
:)
Red Hat's continued virtual monopoly of the enterprise space has made them arrogant to a degree, and a little lazy. Having Novell around gives us a check against them.
That, and SLES just rocks over RHEL 3
What do you mean by better than PAM? What kind of problems are you seeing with PAM ?
Ok, Red Hat have their RHN solution. They also provide a tool named Satellite which effectively provides RHN on your own site. The idea is that it will manage your entire environment from deployment, through patching, rollback, configuration management, etc. On paper and in the demos it looks like an amazing product.
:)
:)
Having said that, it's looking increasingly likely that this pile of shit will cost me my job. From small problems like being able to take the entire Satellite solution down with a single security scan, to half implemented features, to poor support, I'm currently fighting a rear guard action trying to justify this product where I recommended it. You can't have multiple tiers with the overview information all being visible at the top of the tree like you can with SMS, etc.
I've learnt one very valuable lesson from all of this - People who laugh at people who say no-one ever got fired for buying Microsoft, obviously haven't been on the firing line for buying something else.
Ignoring Satellite, that pretty much leaves the following:
1. Their support period - they guarantee to support the distro for 7 years now
2. Guarantee of API / ABI stability - This is great on paper, but doesn't happen. It didn't happen in RHEL3 because of the changes to Linux Threads, but a simple fix with LD_ASSUME_KERNEL was available. Still, it scared managers. I have no idea how they plan to do this in RHEL4 because of the new kernel development process with so much change happening in the 'stable' kernel.
3. Support offering - so far, IMHO, this is shit. I've had such gems as being told that I shouldn't hot-swap disks in an HP-DL380, but that I should power the machine down and replace the disks. I'm so glad I have that in writing - makes for great stories at parties
4. SELinux in RHEL4 - This looks good, but I think it will be a while before enterprise apps such as Oracle RAC work seamlessly with this.
5. AD integration - They are providing out of the box AD integration with RHEL4 but it's pretty primitive. It just uses Winbind and this isn't suitable for a lot of bigger oranisations with more complex AD heirachys.
Overall, I'm not impressed with Linux in the Enterprise. Having dealt with Red Hat and Novell, even though I'm a dyed in the wool Linux geek, I still feel more confident about Solaris. Even with Sun's schizophrenia
I've written to every one of my MEP's and got responses from 2 of them.
The response from the Liberal Democrat was quite good. She sits on the JURI council who voted to restart patents, and isn't happy with the outcome. She was quite optimistic several weeks ago when I started corresponding with her, but this has changed.
The FastFood^WNew labour candidate is fully in support of them, so I'm casting a punishment vote in the upcoming general election.
What about countries where English isn't that common? In those places, MS will have to run local call centres and that will push their costs up. Surely this increase in costs will be more than the savings from thwarting privacy ?
Red Hat are probably the most closed company I've ever seen. They may still do a lot of development and contribute in code and investment (hardware certification, etc.) but at a personal level, I'd rather deal with Microsoft than with Red Hat.
:)
When Windows server 2003 launched, I needed to get familiar with it, so I went to Microsoft's site and downloaded a 120 day eval copy. Using this, I was able to get familiar with the changes, etc.
When RHEL3 came out, I needed to do the same. Now, I'm an RHCE, so I've paid them a lot of my money directly (600 quid just for the exam), and sent even more money their way by recommending Red Hat wherever I went.
Do you think I could get an eval ? Fuck no! I called the Guildford office in London, I called the head office in the USA, I e-mailed, I tried everything.
Two months later, I had an interview at an investment bank, and they asked if I had experience with RHEL3. I explained that I didn't, but I had experience with RHEL 2.1AS as well as their free versions.
Microsoft 'get it' - they realise that if they help me learn their products, I can get a job using their products and they will sell more of their products. And I've never given MS a penny for training or certification.
Red Hat ? This bastion of freedom, this shining light of openness won't let me trial their products - they won't give you a thing to help you unless you pony up more money.
Sure, I could have spent the $200 on RHEL3 workstation, but with that attitude why should I? SuSE were quite happy to provide a trial version of SLES and now that's what I sell to all of my clients. I don't lose out, SuSE gains, and Red Hat lose out on a couple of enterprise contracts. It's all good
Just move to calling it Dead Rat - anyone who's ever tried to get support from them will immediately know what you mean :)
And no, I'm not bitter because I wasted my saturday because they shipped a dodgy certificate for the Satellite server. No sir, not me.
Practically nobody? You mean nobody who uses x86 technology. Sun Ultra Sparc 4 chips have been dual core for quite some time now. I believe that IBM also has dual cores in their power series.
:)
X86 is not the whole world
Exqueeze me please?
:) So not only are you getting the processor cheaper, you're HALVING your licence costs.
Everything I've read so far says that two separate chips will give better performance than a dual core at the same clock speeds.
So if you have a dual Xeon 3.6Ghz, you're likely to get better performance than a machine with a single dual core 3.6Ghz.
This comes down to cores having to wait for access to resources, etc.
This is why I don't like the dual core == dual licence scheme. I'm _NOT_ getting twice the performance as with a single chip, but I have to pay twice.
In fact, this is something that makes Fujitsu servers attractive as competition for Sun. You can get equivalent performance to a dual core Sun Sparc IV 1.25Ghz with a single 1.8Ghz Fujitsu Sparc processor. Those clock speeds might be slightly out, but find the nearest
Remember, it's not just a few players in the enterprise market that licence like this. Veritas, Oracle, HP Openview, Websphere MQ, they all do this. So if you can get the same performance from a single core CPU as you can from a dual core, halving your licence costs can be a big deal!
Hang on a second! You mean if we get rid of beastie, we won't have hot women dressing up in tight red leather anymore?
:D
SIGN THAT PETITION DAMMIT!
There are many cases of Microsoft deviating from accepted standards. The reasons that are normally given for this generally don't stand up to public scrutiny.
Why does Microsoft persist in breaking standards just to lock the competition out? Is it that you are too scared to compete on the merits of your products, the fact that this behaviour has become institutionalised, or some other reason?
Winbind is worthless for most real purposes. You cannot assign a UID to a user and there is no guarantee that a user will have the same uid each time they log on.
If I log on now, and get a UID of 1000, then at some point changes are made to the AD server, or a new version of samba is installed that supports computer accounts, when I next log-in, I will not get that same UID.
There are also issues with groups.
LDAP can be used quite effectively, but again there are limitations and issues. I have got around most of these, but not all of them.
Who said anything about the race of the professional car thieves ? If you scratch beneath the surface you'll find more than just a few of the syndicates are white owned and controlled.
So what does the lesson become then? That the Brits should have won the boer war and kept those uppity farmers in their place? Yeah - that sounds reasonable. It's not like a British subject was recently convicted in South Africa for participating in the planning of a coup. Oh, wait it is.
Maybe then the moral should be that you can't blame entire demographics for _anything_ and that you actually need to look at individuals.
But that would be like, work man!
Sheesh!
In some countries, car theft is not just something that happens occasionally - it's an industry. And as in all industries, there are the rank amatuers and the pro's - For the pro's, this looks like a good option.
:)
:)
Consider South Africa - an entire arms race grew up around car theft. First the thieves just took cars when they were parked, so the insurance companies insisted that everyone have alarms and immobilisers.
The thieves got around those pretty quick - rumour is that a lot of professional's signed up for work at installation centres, learnt their way around them, and went back to work.
Next step was the gearlock - a device that locks the gearstick into a specific gear. IIRC, you couldn't remove the key on the earlier units unless you had the gearlock in, and if your car was stolen, the insurance company insisted on seeing all 3 keys.
Now with cars being so hard to steal, the age of the hi-jack was ushered in. If they can't get your car while it's parked, they'll take it while it's roll rolling.
In response, anti hi-jack systems became the norm. I can't remember how it was activated, but basically the bad guys show up, you let them take the car, they roll 20 metres down the road and the car cuts out and an alarm starts going off.
Around the same time we also go Satellite tracking, although I seem to remember something about it actually using the cellular infrastructure (GSM) not satellite - I may be wrong on this. Initially, the recovery rate on stolen and hi-jacked cars went through the roof. Unfortunately, the bad guys just upped the stakes. Soon we started seeing more kidnappings and murders as part of hi-jacks because if you can't call the stolen car in, they have longer to chop it.
Many vehicles were stolen to order, and not just new cars. Older cars that were common on the road were often targetted, then broken for spares. Cars that you wouldn't normally think twice about were stolen for export to Botswana and Zimbabwe, because the availability of spares for these made them popular vehicles.
Of my close circle of family and friends, we have had at least 10 cars stolen. Of those, not a single one has been recovered, so it's not a huge risk occupation really
I'm willing to bet that if this flaw is used anywhere, it will be used in South Africa - it's just one more tool for the biggest growth industry around
Sun says that customers are demanding it, but I wonder how real that demand is...
I know for a fact that Sun have sent letters to some customers asking them to ask ISVs about support for solaris 10. From what I can tell, this is to try and create the illusion of a grassroots movement where none exists.
So in light of this, how much of this 'demand' is actually real, and how much of it is Sun customers doing what Sun tell them in the hopes of a more favourable relationship when their support contract comes up for renewal ?
Of course, if you have an iPod, you could just buy iLingo which is a portable phrase book for the iPod. It gives many phrases and apparantly will speak them for you as well.
My wife bought an iPod purely because of this application!
Doing the Right Thing(tm) in politics these days is a career death sentence more often than not.
If a lawyer worked this out, consider it patented already.