I don't believe there's much of a computing market between something you can fit in your pocket and something quite heavy to carry around. Once i've got to carry a bag to hold something, I may as well carry a laptop that gives me much more power.
If it doesn't fit in my pocket, it's not something I can just take out and use, like in an airport or railway station. In which case, I have to take something out of a case and sit down with it. May as well just use my laptop that has a keyboard and costs a lot less.
Webapps are key to this. People are deploying them more and more. Not to create a platform-agnostic open environment, but because they typically don't need a bunch of testing in terms of impact on the rest of the desktop.
Creating less dependency on the desktop is a nice spin-off.
I hear where you are coming from, and I'm not dismissing RoR out of hand. I'm just saying that sometimes frameworks can be very seductive, and that demos can often make things appear as a shortcut.
The key is applications, and reducing your dependency on Windows-only applications. Web-based apps are one part of this. The other is getting onto OpenOffice.org.
Once you've got people running happily on those, you can then migrate people.
I worked on mainframe 4GL systems nearly 20 years ago that could do screen to record mapping. Likewise, MS Access has had the same functionality for a long time.
The point is that ActiveRecord type technology is nothing new, nothing revolutionary. It's also very seductive, until you actually start using it for real, and find that in a lot of environments you don't simply do a table read/table update for a lot of applications. You often need to take a view of data across many tables, or do updates that are more complex.
That's a very sensible way to think about it. Unfortunately, it's rarely the way that large corporates think (but small companies who outsource things like email do).
I recently helped a small company set up their staff on a remote Exchange server. A little slow, but it gave them all the Exchange functions. It costs them a fixed price per user per month of about $10 and they get more storage than most companies hand out.
They don't mind paying out $10/user/month, so why do most companies want to do email on the bloody cheap, and impose insane limits like 50mb/user? How much is your staff's time worth to you?
How about you actually go out and work in a marketing department, or do some software project management, and see how insanely frustrating email limits are. How expanding email services would actually make the company more productive.
I've gone to send an email to a client, only to be told that my email is over the limit. So, I have to save the draft, and go into my inbox. I then delete any "funnies" that someone might have sent me. Still doesn't work. So, now, what do I do? File my email in a "local" folder that sits on my hard drive. How much did that just cost the company instead of spending money on extra storage?
If you've worked as a project manager, you know that every email related to the project is important. The golden rule is to delete nothing until at least the project has been implemented, preferably delete nothing ever to do with that project. You never know when something might be important.
Why isn't email a good place? Personally, I like email. If I want to see a customer conversation, it's all there. Dated, with context and everything else.
If I copy a document sent to me to somewhere else, where does the context go?
Email limits are rarely desired by anyone in the business and typically get in the way of the money making of the business. In my experience, they force expensive people to have to waste their time shifting mail from mailbox to other places.
Compared to someone having to backup 450GB, it's a big expense.
I do pieces of occassional work in Access, and I'm still doing it in Access 2000. In part, because I know that everyone will have that version or better, but also because there's been little good added since.
Anything that's new already has been dealt with by 3rd party products, or developers finding another way.
Someone might say "but now it's built in!", but so what? If I'm geared up with 3rd party greatness already, why do I want to spend time switching to something else?
Lots of large companies are still running Office 2000 for 2 reasons:-
1. There's nothing worth having since then.
2. Switching people costs money and time.
I used Office 2003, and all I got was something that impaired me, because my brain was trained for Office 2000, and o2k3 broke that. And, I used no new features.
There's one area where Office still wins - Base sucks compared to Access. I also don't do much powerpoint, so can't comment there.
But for me, Writer does what I need to produce detailed technical documents. Calc does all I need for a spreadsheet (which isn't that much, but then I imagine it's more than most people need).
I'd like to know where it is "leaps and bounds" ahead of OOo.
In a lot of large corporations, they just deal with Office in a similar way to "single isolated users". My experience is that the collaboration features are largely ignored in favour of simple emailling.
Without a doubt, there probably are people out there who got to be first in line for a leather mac case, but the initial coverage was very much "leather case? boombox? what a waste of time. Jobs has jumped the shark. Apple suck now. Oh, and by the way, they also released an intel mini".
I'm a PC user and the mac mini is a big deal. The fact is that I've been worried about getting a mini because of the performance, but I would be a lot more confident about getting one now, although I'd like to see some sort of benchmarks of real performance (ie not just chip speed comparisons).
And what percentage of home PC users actually care that much about games? As in, care about anything beyond solitaire or a few retro games? I certainly know a lot of over 30s who couldn't care less.
There's a quote by Victor Hugo that goes something like "nothing, not all the armies of the world can stop an idea whose time has come".
Often, there are ideas out there, which for various reasons just don't work. For instance, Martin Luther's ideas for the reformation were probably explored by people before. But, with the invention of the printing press, and being able to communicate to the masses, it could come to fruition.
Richard Stallman has been talking about free software for how long? So why did it take so long? Because, until there was mass-market high-speed internet, communicating information about, and ultimately being able to download 600+MB of data was a serious undertaking.
Two things are not going back - open source and web-based applications. Neither are fads, and are going to grow more, because the things required to make them work are growing, as are people's needs in terms of communication changing.
I've used some OSS code, and made changes and offered those changes back. Even though technically, I didn't have to give those changes back, I did, and for two reasons...
1. Because it's the right thing to do.
2. Because if other people are using my changes, it means that bugs might get spotted. It also means that any revisions to the base code also include my changes, so I don't have to reapply them.
As well as community reasons, there are plenty of good commercial reasons for using and contributing to FOSS.
Sounds to me like Ellison is right. Watch your competitors, large and small.
However, he's also right that they are mostly unaffected by MySQL. MySQL is a good product, but right now, it ain't Oracle. And some people are prepared to pay the sort of money that Oracle can cost.
Slashdot Mirror
I don't believe there's much of a computing market between something you can fit in your pocket and something quite heavy to carry around. Once i've got to carry a bag to hold something, I may as well carry a laptop that gives me much more power.
If it doesn't fit in my pocket, it's not something I can just take out and use, like in an airport or railway station. In which case, I have to take something out of a case and sit down with it. May as well just use my laptop that has a keyboard and costs a lot less.
Creating less dependency on the desktop is a nice spin-off.
I hear where you are coming from, and I'm not dismissing RoR out of hand. I'm just saying that sometimes frameworks can be very seductive, and that demos can often make things appear as a shortcut.
Once you've got people running happily on those, you can then migrate people.
It's Google News with an extra side order of suck thrown in.
no, I'm thinking more about retrieving a much more complex data query than that.
I worked on mainframe 4GL systems nearly 20 years ago that could do screen to record mapping. Likewise, MS Access has had the same functionality for a long time.
The point is that ActiveRecord type technology is nothing new, nothing revolutionary. It's also very seductive, until you actually start using it for real, and find that in a lot of environments you don't simply do a table read/table update for a lot of applications. You often need to take a view of data across many tables, or do updates that are more complex.
So don't deal with them.
I recently helped a small company set up their staff on a remote Exchange server. A little slow, but it gave them all the Exchange functions. It costs them a fixed price per user per month of about $10 and they get more storage than most companies hand out.
They don't mind paying out $10/user/month, so why do most companies want to do email on the bloody cheap, and impose insane limits like 50mb/user? How much is your staff's time worth to you?
I've gone to send an email to a client, only to be told that my email is over the limit. So, I have to save the draft, and go into my inbox. I then delete any "funnies" that someone might have sent me. Still doesn't work. So, now, what do I do? File my email in a "local" folder that sits on my hard drive. How much did that just cost the company instead of spending money on extra storage?
If you've worked as a project manager, you know that every email related to the project is important. The golden rule is to delete nothing until at least the project has been implemented, preferably delete nothing ever to do with that project. You never know when something might be important.
Given the choice of having an email locally, or deleted, I know which I'd choose.
If I copy a document sent to me to somewhere else, where does the context go?
Compared to someone having to backup 450GB, it's a big expense.
That's Microsoft's problem in a nutshell. There's basically little to add to MS Office and at the same time, OOo is adding features.
Anything that's new already has been dealt with by 3rd party products, or developers finding another way.
Someone might say "but now it's built in!", but so what? If I'm geared up with 3rd party greatness already, why do I want to spend time switching to something else?
1. There's nothing worth having since then.
2. Switching people costs money and time.
I used Office 2003, and all I got was something that impaired me, because my brain was trained for Office 2000, and o2k3 broke that. And, I used no new features.
But for me, Writer does what I need to produce detailed technical documents. Calc does all I need for a spreadsheet (which isn't that much, but then I imagine it's more than most people need).
I'd like to know where it is "leaps and bounds" ahead of OOo.
In a lot of large corporations, they just deal with Office in a similar way to "single isolated users". My experience is that the collaboration features are largely ignored in favour of simple emailling.
A lot of girls don't get into those stupid contests, like a lot of guys don't get into contests about largest, stupidest, fastest car.
I'm a PC user and the mac mini is a big deal. The fact is that I've been worried about getting a mini because of the performance, but I would be a lot more confident about getting one now, although I'd like to see some sort of benchmarks of real performance (ie not just chip speed comparisons).
And what percentage of home PC users actually care that much about games? As in, care about anything beyond solitaire or a few retro games? I certainly know a lot of over 30s who couldn't care less.
Often, there are ideas out there, which for various reasons just don't work. For instance, Martin Luther's ideas for the reformation were probably explored by people before. But, with the invention of the printing press, and being able to communicate to the masses, it could come to fruition.
Richard Stallman has been talking about free software for how long? So why did it take so long? Because, until there was mass-market high-speed internet, communicating information about, and ultimately being able to download 600+MB of data was a serious undertaking.
Two things are not going back - open source and web-based applications. Neither are fads, and are going to grow more, because the things required to make them work are growing, as are people's needs in terms of communication changing.
I've used some OSS code, and made changes and offered those changes back. Even though technically, I didn't have to give those changes back, I did, and for two reasons...
1. Because it's the right thing to do.
2. Because if other people are using my changes, it means that bugs might get spotted. It also means that any revisions to the base code also include my changes, so I don't have to reapply them.
As well as community reasons, there are plenty of good commercial reasons for using and contributing to FOSS.
However, he's also right that they are mostly unaffected by MySQL. MySQL is a good product, but right now, it ain't Oracle. And some people are prepared to pay the sort of money that Oracle can cost.