If you use strong crypto only for some connections, this is would reveal your communication patterns. This may show who your fellow terrorists are. That way homeland security could place a bug in their office. Or just send all of you Guantanamo just to be on the safe side.
It's not more vulnerable with respect to legal matters. It might be financially more vulnerable though as opensource projects can't afford to fend themselves even against bogus patent claims. Open sorce projects often have less patens to use for cross licencing.
In many European countries the situation is somewhat better than in the US as the loosing part in a trial pays the legal fees for both parties. That might make it less tempting to make bogus patent claims.
"Having said that, KDE has way more bugs/quirks than GNOME but its still easier to use."
If its easier to use or not largely depends on what you use it for and who you are. If you don't know who the intended users are and their needs you will not be able to build a good system.
If you read the KDE usability list and various KDE development lists, you get the impression that the usability experts and the people that writes the code are working on two different systems.
The usabilty experts tryes to build a system that anybody including my mother can use, while most developers seam to target the needs of advanced usrs at least when the usabilty people doesn't manage to talk them out of it.
I'm not saying its wrong to target advanced users, but if you do, you should do so consistently. The risk is that we end up with systems that is dumbed down in some areas while the it is still too hard to use for some users. End result, it will be hated both by noobs and superusers.
The Gnome people seam a lot more focused at the moment. The KDE people need to make up their mind, or they will fall behind. This would be a pity since the KDE/QT framework is really slick.
I would say that KDE is in much more need of a fork than Gnome. A fork of KDE that was as simple to use as Gnome, or perhaps even looked like Gnome but built on QT technology. It would be unbeatable.
Then the current KDE developers could continue as usual an focus more on advanced users.
2 days? Isn't that a very long time for finding the advanced preference where you turn off mozillas ability to hide the toolbar, statusbar, changeing the statusbar text, moving or rezizing existing windows, etc
Seriously, this is hard to fix. Some people need these features. Just turning them off by default is not a good solution.
Perhaps they could turn them off by default, and then allow users to turn them on as they occur on a site by site basis . They already have a similar solution for popup windows.
If software patents, gets granted like they are in the US today, where every simple and obvious idea seam to be granted usually overly broad patent regardless how much prior art there is, the future of the software industry belongs to the lawyers.
Given the amount of such bogus patents floating around I doubt that it is possible to write any software longer than 1000 lines of code without infringeing on at least one existing patents.
Now I'm just waiting for sombody to file a patent on the procedure of filing bogus patents and licence them at slightly lower cost than it would cost to contest them in court.
Software patents threatens all software development and innovation in all software industry not in only free software. Free software is however somewhat more vulnerable as it is impossible to solve patent issues by licence fees.
The only option for free software would be to trade patents, but the problem is that few free sofware developers do patent their stuff. So there will be little to trade.
Besides, how do we handle companies buying or seeking patents on technology without having any software business on ther own. Companies just sitting on patens waiting for some patented technology to become a taxable hit. Not even companies like IBM would be able to fight them by using their enormeous list of patents as a weapon, as they could if the enemy was an ordinary software producein company.
You are right ASP and ASP.NET are two entirely different beasts.
But this is probably more of a problem than an advantage. Many people that once was creating (bad?) ASP scrips did not have a programmer background.They were content creators that with some difficulty struggled to get their HTML code scriptable.
These developers still exists, and are very common. They will have a very hard time switching to ASP.NET or for that matter make good use of the new object oriented features in PHP5.
But if you are not a programmer, and need or can make use of all the fancy features of ASP.NET or PHP5 you don't need to consider what platform has the technical edge. You would consider what platform is most widely used as that would give you better return on your investment in learning the language.
So far PHP runs well on both Apache and IIS, while Mono is very untested on Apache/Linux. Given that Apache runs on 2/3 of all internet servers the PHP would be a better option for this type of people.
Like you, many people complain about the install process, in Linux and other free unix clones. They almost always mention the./configure, make, make install procedure.
But what you describe is the build process not the the install process. If you consider this part of the instll process, the same critisim could be raised against any OS, including windows.
Using modern install tools from the open source world like yum, aptget or urpmi. Dependencies are resolved autogmagically and downloaded over internet if needed. The user is not bothered by difficult questions like where he wants to install the software.
You just do yum install programname and the desired functionality is there for you to use. It really can't get more simple.
Perhaps it even is a bit too simple. It kind of reminds me of old audio cassettes that was shaped to automagically tell the cassette deck of what type of tape they contained. Yet almost all advanced cassete decks had a switch where the user could select the tape type manually. Why? The simple answer was that setting the tape type manually gave users a feeling of control and expertise in the audio domain.
This is why your Linux using friends tells you about how they use tar,./configure, make, etc to install their programs. Just telling you that they installed mozilla by selecting it from a GUI in urmpi or even writing yum install mozilla will not make him feel like a unix guru nor would you recognize him as such.
One of the reason for choosing GPL would be to control competition, as it eliminates the risk that some of your competitors adds new features that you have no access to.
I have always regarded BSD like licences suicidal if you issue and only beneficial to the licencee. While GPL gives more equal terms at least on paper. In reality the parti that have written the major part of the code will probably come out on top as he will have better understanding on how it works and will probably be able to provide better services.
So given Microsoft normally highly competitive behavior, one wonders if their hate towards GPL have clouded their minds.
Switching browsers browsers is not enough. Who knows, Mozilla could be the target of some malware tomorrow. Switching to Mozilla just buys you some time.
To be more secure we need an OS that prevents the browser from executing unauthorized code and prevents the browser from accesing sensitive information or applications on our systems. The browser should not be allowed to be the only layer of security.
One way would be to swich to some Linux, using a distro that make use of the SELinux stuff enables mandatory access control and set up a good security policy.
But who will be willing to pay for it. I guess that much of the existing space shuttle technology have been developed with military purposes in mind. That would probably make governments more willing to put money into conventional technology in the name of national security, than into this new and probably more vulnerable elevator.
Another question is, will it really work? Even if each fibre in cable could be almost unbreakable, each fibre need to be very long or you would have a high probability of having all fibres end in one specific place along the cable, creating a weak spot. From what I understand they have had this kind of trouble in deap sea expiditions, and in this project we are talking about much longer distances.
If I took a Unix course back in 1989 (before Linux even had emerged) most of what I learnded then would still be somewhat useful in Linux of today. How much would 15 years old windows knowledge help me in manageing windows XP of today. Not much I think. Most likely I would have to have more frequent retraining if I run windows.
I don't think windows is particularly easy to use or set up. The only reason people use it is that it comes with the computer. Here an example from last week:
I was trying to install win2k server on my wifes computer. It took at least 10 hours including service pack downloads over a ADSL connection. It was supposed to dual boot together with an NT4 workstation.
What happened. It was impossible to get it to recognize the Nokia 19" screen connected to a Matrox card. The screen just went totally black unless using plain old 640x480 vga mode. I guess that it got the wrong sync freqency. It also did something to the existing NTFS partitions so that they could not be read by NT4 anymore.
The soluton was to put in a Knoppix CD to recover the files. It booted nicely. It identified the screen and graphics card correctly and gave a resonable screen resolution (1024x1280 I think), it also identified the flascard reader and the soundcard automagically. And above all it could read the diskpartitions that win2k had rendered unreadable to NT4 and it made it possible to copy the files to another NT4 server using samba.
Now tell me whats most easy, spend 10 hours and not get windows work, or a minute or two to boot Knoppix. And Oh, I forgot, Knoppix came with OpenOffice and lots and lots of other productivity software that probably had taken a couple of hours more to install to get similar functionality in windows.
As I see it, there is no reason to have to cope with DRM technology just because windows is so much easier than the alternatives.
Actually, I think MS will have a large problem suing mr Amadeu. They can afford the lawyers and the best judge money can buy, but they can't afford the press coverage such a lasuit would generate. Many people share the views of mr Amadeu and would probably be on his side, and become even more hostile to Microsoft.
Even if Microsoft manages to buy some of the press, there is a significant risk that some website like groklaw may emerge and start digging up annoying facts on Microsoft and its business practices.
The question is will Microsoft be smart enough to realize this.
Evolution and Korganizer are nice calendar apps, but to my knowledge they are not available on windows.
A cross platform calendar app is very much needed or it will slow down the adaptaion of free desktop alternatives such as Linux.
Why is that? Managers need to do the switch to gradually, and as more and more applications becomes available for the free platform switch the entire organization. Calendaring is often a mission critical application and if managers have to support two different system while they are migrating, the Linux migration will be more costly.
So, the mozilla calendar is extremely important. It's so important that I would even advice the Gnome people to either port Evolution to windows or drop it in favor of a Mozilla solution.
Even though Evolution is very good, I guess that dropping it from Gnome would be the best thing to do, as if they do port it to windows it will have to support every feature of MS-Outlook that it tries to clone anayway, or users will be disappointed. It will not matter if it have other additional advantages over MS-Outlook, windows users will just compare what MS-Outlook can do to what a windows based Evolution can do. We have already seen this phenomena when trade magazines compare OpenOffice to MS-Office.
So going with Mozilla for Calendering would probably be a tactical thing to do. However to do so it will need a lot of development as it is rather weak as it is.
Code ode should not need to be optimized for a specific compiler. As long as programmer time is more expensive than faster hardware, code should be optimized for readability and maintanabilty not for speed, unless you program for very extreeme conditions.
Hopefully both designers and programmers. It would be pointless otherwise. People in the role of designers create it, and people in the role of programmers read it. Somtimes the different roles is held by the same person somtimes it is not.
You may be right that programmers can express themselves in greater detail in a programming language. But just because eskimos have more than 50 different words for snow it doesn't mean that they can communicate well with other people with less expert knowledge in that field.
It about communication much more than about expression. And UML allows you to create model of different level of detail to be used by different persons and at different stages in the development process. From the customer that will need to understand the system at least on a conceptual level to the programmer that have more needs for detail.
I fail to see why you think designers should be less competent than programmers. Even though UML can be used to describe code, the real benefit is not mapping of code into UML diagrams, but rather to capture a business reality into a readable conceptual model that we can elaborate on and refine to use as a base for creation of software artefacts
Re:Designs cast in Stone
on
UML Fever
·
· Score: 1
Story boards, Personas etc, is very good for catching how to design userinterfaces. But designing a user interface is quite different from designing an application. I certainly agree that the interactionn with the end user must work to make the application a success. But it is not the only thing needed.
First of all, without a good conceptual model of whatever business setting the application is supposed to live in and what concepts that is supposed to be part of the system, it is harder to communicate with the customer and to see his goals with the application. It is also very easy to fall into the trap of programming existing business processes and fail to see possible ways to simplify it.
You would hardly go to a building contractor and ask him to build you a house based on a storyboard on how you intended to use it. In such a house you would probably be able to move from the bedroom to the bathroom just like you specified, but you could easily fail to recognize that it would be a good idea to have the kitchen and the bathroom only seperated by a wall to reduce the costs for plumbing. This why you hire an architect to make a drawing of your future house before you build it. In the software world that drawing would be the UML diagram.
And just like in a building project we will have drawings/UML-diagrams with different detail aimed at various persons in the building process.
Once a conceptual model is established we can start think of how to transform it into software artefacts creating new UML diagrams where we dress up our concepts to form the functionality of our system.
Then you connect that functioality to your GUI designed by the methods you describe.
In this design process creating UML diagram not only makes communication easier, but it also makes the various steps in the design process more evident. This increases the chance that the right design decisions are taken by the right people and at the right time in the process.
If you only design with user inteface in mind chances are that your system will be hard to maintain in the long run.
Groups that all sound alike, and shaped in a form that people are supposed to like is all there is. No thanks I'm not interested.
And even if there was something I liked I would think twice before buying. There are so much non standard, copy protected CDs around that may or may not be playable on my equippment, and I'm not taking the risk that they won't play.
Besides the fact that the CDs are copy protected, sort of implicitly accuses me of being a thief. I don't like that, or at least not trustworthy. So they can keep their music.
My guess is that it would take a week less time to have your developer up to speed with QT than what it would take if they used e.g. GTK. And there you have your $1000. Assume that a developer costs you $50/h and that he works 40h/week. That would give you a developer cost of $2000 in that week. This means that using QT have saved you $1000 in just one week. Your $1000 is probably less than what you pay for one sinle ad in a the press to get your application sold.
This line of reasoning is based on that QT actually is easier to use. Why do I think this is the case? Well, I was able to fix bugs and do minor application development after 3 hours in QT but in GTK I don't even know where to start. It could be that QT is better documented, but fact remains.
And if your application idea isn't unique enough so that it can't stand the competition from free sofware, then perhaps you should develop it as free software and and make your money on services related to it, or invent something better.
No not really. You can licence it through GPL, if you do open source work, or you can buy licences from Trolltech at a reasonalble price if you want to do closed source work.
The fact that you have to buy a licence for closed source may be a problem for hobby developers that have more time than money. To companies like Novell the quality is more important than the price . To them, time to market is everything. And they will spend far more money on developer saleries and marketing than they will spend on QT licences. So they will go for the best tool. And QT is one of the best there is. (Cross platform, good development tools, well designed framework...)
And if they want to release free software they have less risk of being hurt by some competitor if they release it under GPL than under LGPL or BSD type licences. I guess this was the reason MySQL AB changed the licence on their client libs to GPL.
"security: OOo provides many of the same security features as MSO, but some of them are not needed for OOo">
Microsoft could actually be right when they claim that their Office suite have better security.
This is because it supports TCPA. But to do that additional Microsoft software would be required.
As this software runs on the server side it could also create a single point of failure, You would also almost certainly have to do a lot of eductation to both ordinary users and administrators.
But security should not be something that resides at the application level. The OS should take care of such things. So if TCPA doesn't work with OOo we can only conclude that TCPA is flawed. Not that it is something wrong with OOo.
Slashdot Mirror
Actually, it does.
If you use strong crypto only for some connections, this is would reveal your communication patterns. This may show who your fellow terrorists are. That way homeland security could place a bug in their office. Or just send all of you Guantanamo just to be on the safe side.
It's not more vulnerable with respect to legal matters. It might be financially more vulnerable though as opensource projects can't afford to fend themselves even against bogus patent claims. Open sorce projects often have less patens to use for cross licencing.
In many European countries the situation is somewhat better than in the US as the loosing part in a trial pays the legal fees for both parties. That might make it less tempting to make bogus patent claims.
"Having said that, KDE has way more bugs/quirks than GNOME but its still easier to use."
If its easier to use or not largely depends on what you use it for and who you are. If you don't know who the intended users are and their needs you will not be able to build a good system.
If you read the KDE usability list and various KDE development lists, you get the impression that the usability experts and the people that writes the code are working on two different systems.
The usabilty experts tryes to build a system that anybody including my mother can use, while most developers seam to target the needs of advanced usrs at least when the usabilty people doesn't manage to talk them out of it.
I'm not saying its wrong to target advanced users, but if you do, you should do so consistently. The risk is that we end up with systems that is dumbed down in some areas while the it is still too hard to use for some users. End result, it will be hated both by noobs and superusers.
The Gnome people seam a lot more focused at the moment. The KDE people need to make up their mind,
or they will fall behind. This would be a pity since the KDE/QT framework is really slick.
I would say that KDE is in much more need of a fork than Gnome. A fork of KDE that was as simple to use as Gnome, or perhaps even looked like Gnome but built on QT technology. It would be unbeatable.
Then the current KDE developers could continue as usual an focus more on advanced users.
2 days?
Isn't that a very long time for finding the advanced preference where you turn off mozillas ability to hide the toolbar, statusbar, changeing the statusbar text, moving or rezizing existing windows, etc
Seriously, this is hard to fix. Some people need these features. Just turning them off by default is not a good solution.
Perhaps they could turn them off by default, and then allow users to turn them on as they occur on a site by site basis . They already have a similar solution for popup windows.
If software patents, gets granted like they are in the US today, where every simple and obvious idea seam to be granted usually overly broad patent regardless how much prior art there is, the future of the software industry belongs to the lawyers.
Given the amount of such bogus patents floating around I doubt that it is possible to write any software longer than 1000 lines of code without infringeing on at least one existing patents.
Now I'm just waiting for sombody to file a patent on the procedure of filing bogus patents and licence them at slightly lower cost than it would cost to contest them in court.
Software patents threatens all software development and innovation in all software industry not in only free software. Free software is however somewhat more vulnerable as it is impossible to solve patent issues by licence fees.
The only option for free software would be to trade patents, but the problem is that few free sofware developers do patent their stuff. So there will be little to trade.
Besides, how do we handle companies buying or seeking patents on technology without having any software business on ther own. Companies just sitting on patens waiting for some patented technology to become a taxable hit. Not even companies like IBM would be able to fight them by using their enormeous list of patents as a weapon, as they could if the enemy was an ordinary software producein company.
You are right ASP and ASP.NET are two entirely different beasts.
But this is probably more of a problem than an advantage. Many people that once was creating (bad?) ASP scrips did not have a programmer background.They were content creators that with some difficulty struggled to get their HTML code scriptable.
These developers still exists, and are very common. They will have a very hard time switching to ASP.NET or for that matter make good use of the new object oriented features in PHP5.
But if you are not a programmer, and need or can make use of all the fancy features of ASP.NET or PHP5 you don't need to consider what platform has the technical edge. You would consider what platform is most widely used as that would give you better return on your investment in learning the language.
So far PHP runs well on both Apache and IIS, while Mono is very untested on Apache/Linux. Given that Apache runs on 2/3 of all internet servers the PHP would be a better option for this type of people.
Like you, many people complain about the install process, in Linux and other free unix clones. They almost always mention the ./configure, make, make install procedure.
./configure, make, etc to install their programs. Just telling you that they installed mozilla by selecting it from a GUI in urmpi or even writing yum install mozilla will not make him feel like a unix guru nor would you recognize him as such.
But what you describe is the build process not the the install process. If you consider this part of the instll process, the same critisim could be raised against any OS, including windows.
Using modern install tools from the open source world like yum, aptget or urpmi. Dependencies are resolved autogmagically and downloaded over internet if needed. The user is not bothered by difficult questions like where he wants to install the software.
You just do yum install programname and the desired functionality is there for you to use.
It really can't get more simple.
Perhaps it even is a bit too simple. It kind of reminds me of old audio cassettes that was shaped to automagically tell the cassette deck of what type of tape they contained. Yet almost all advanced cassete decks had a switch where the user could select the tape type manually. Why? The simple answer was that setting the tape type manually gave users a feeling of control and expertise in the audio domain.
This is why your Linux using friends tells you about how they use tar,
One of the reason for choosing GPL would be to control competition, as it eliminates the risk that some of your competitors adds new features that you have no access to.
I have always regarded BSD like licences suicidal if you issue and only beneficial to the licencee. While GPL gives more equal terms at least on paper. In reality the parti that have written the major part of the code will probably come out on top as he will have better understanding on how it works and will probably be able to provide better services.
So given Microsoft normally highly competitive behavior, one wonders if their hate towards GPL have clouded their minds.
Why holding David Boise against him? The more work that lawyer gets the better. That way he will have less time for other high profile cases.
Switching browsers browsers is not enough. Who knows, Mozilla could be the target of some malware tomorrow. Switching to Mozilla just buys you some time.
To be more secure we need an OS that prevents the browser from executing unauthorized code and prevents the browser from accesing sensitive information or applications on our systems. The browser should not be allowed to be the only layer of security.
One way would be to swich to some Linux, using a distro that make use of the SELinux stuff enables mandatory access control and set up a good security policy.
As long as I can't order the weather I like where I am from these weather service companies its not worth paying for.
But who will be willing to pay for it. I guess that much of the existing space shuttle technology have been developed with military purposes in mind. That would probably make governments more willing to put money into conventional technology in the name of national security, than into this new and probably more vulnerable elevator.
Another question is, will it really work? Even if each fibre in cable could be almost unbreakable, each fibre need to be very long or you would have a high probability of having all fibres end in one specific place along the cable, creating a weak spot. From what I understand they have had this kind of trouble in deap sea expiditions, and in this project we are talking about much longer distances.
15% higher. Not a chance
If I took a Unix course back in 1989 (before Linux even had emerged) most of what I learnded then would still be somewhat useful in Linux of today. How much would 15 years old windows knowledge help me in manageing windows XP of today. Not much I think. Most likely I would have to have more frequent retraining if I run windows.
I don't think windows is particularly easy to use or set up. The only reason people use it is that it comes with the computer. Here an example from last week:
I was trying to install win2k server on my wifes computer. It took at least 10 hours including service pack downloads over a ADSL connection. It was supposed to dual boot together with an NT4 workstation.
What happened. It was impossible to get it to recognize the Nokia 19" screen connected to a Matrox card. The screen just went totally black unless using plain old 640x480 vga mode. I guess that it got the wrong sync freqency. It also did something to the existing NTFS partitions so that they could not be read by NT4 anymore.
The soluton was to put in a Knoppix CD to recover the files. It booted nicely. It identified the screen and graphics card correctly and gave a resonable screen resolution (1024x1280 I think), it also identified the flascard reader and the soundcard automagically. And above all it could read the diskpartitions that win2k had rendered unreadable to NT4 and it made it possible to copy the files to another NT4 server using samba.
Now tell me whats most easy, spend 10 hours and not get windows work, or a minute or two to boot Knoppix. And Oh, I forgot, Knoppix came with OpenOffice and lots and lots of other productivity software that probably had taken a couple of hours more to install to get similar functionality in windows.
As I see it, there is no reason to have to cope with DRM technology just because windows is so much easier than the alternatives.
Actually, I think MS will have a large problem suing mr Amadeu. They can afford the lawyers and the best judge money can buy, but they can't afford the press coverage such a lasuit would generate. Many people share the views of mr Amadeu and would probably be on his side, and become even more hostile to Microsoft.
Even if Microsoft manages to buy some of the press, there is a significant risk that some website like groklaw may emerge and start digging up annoying facts on Microsoft and its business practices.
The question is will Microsoft be smart enough to realize this.
Evolution and Korganizer are nice calendar apps, but to my knowledge they are not available on windows.
A cross platform calendar app is very much needed or it will slow down the adaptaion of free desktop alternatives such as Linux.
Why is that? Managers need to do the switch to gradually, and as more and more applications becomes available for the free platform switch the entire organization. Calendaring is often a mission critical application and if managers have to support two different system while they are migrating, the Linux migration will be more costly.
So, the mozilla calendar is extremely important. It's so important that I would even advice the Gnome people to either port Evolution to windows or drop it in favor of a Mozilla solution.
Even though Evolution is very good, I guess that dropping it from Gnome would be the best thing to do, as if they do port it to windows it will have to support every feature of MS-Outlook that it tries to clone anayway, or users will be disappointed. It will not matter if it have other additional advantages over MS-Outlook, windows users will just compare what MS-Outlook can do to what a windows based Evolution can do. We have already seen this phenomena when trade magazines compare OpenOffice to MS-Office.
So going with Mozilla for Calendering would probably be a tactical thing to do. However to do so it will need a lot of development as it is rather weak as it is.
Sure they have - UnixWar version 2.0
A long overdue release since the old AT&T/BSD version was utterly out of date.
Code ode should not need to be optimized for a specific compiler. As long as programmer time is more expensive than faster hardware, code should be optimized for readability and maintanabilty not for speed, unless you program for very extreeme conditions.
Hopefully both designers and programmers. It would be pointless otherwise. People in the role of designers create it, and people in the role of programmers read it. Somtimes the different roles is held by the same person somtimes it is not.
You may be right that programmers can express themselves in greater detail in a programming language. But just because eskimos have more than 50 different words for snow it doesn't mean that they can communicate well with other people with less expert knowledge in that field.
It about communication much more than about expression. And UML allows you to create model of different level of detail to be used by different persons and at different stages in the development process. From the customer that will need to understand the system at least on a conceptual level to the programmer that have more needs for detail.
I fail to see why you think designers should be less competent than programmers. Even though UML can be used to describe code, the real benefit is not mapping of code into UML diagrams, but rather to capture a business reality into a readable conceptual model that we can elaborate on and refine to use as a base for creation of software artefacts
Story boards, Personas etc, is very good for catching how to design userinterfaces. But designing a user interface is quite different from designing an application. I certainly agree that the interactionn with the end user must work to make the application a success. But it is not the only thing needed.
First of all, without a good conceptual model of whatever business setting the application is supposed to live in and what concepts that is supposed to be part of the system, it is harder to communicate with the customer and to see his goals with the application. It is also very easy to fall into the trap of programming existing business processes and fail to see possible ways to simplify it.
You would hardly go to a building contractor and ask him to build you a house based on a storyboard on how you intended to use it. In such a house you would probably be able to move from the bedroom to the bathroom just like you specified, but you could easily fail to recognize that it would be a good idea to have the kitchen and the bathroom only seperated by a wall to reduce the costs for plumbing. This why you hire an architect to make a drawing of your future house before you build it. In the software world that drawing would be the UML diagram.
And just like in a building project we will have drawings/UML-diagrams with different detail aimed at various persons in the building process.
Once a conceptual model is established we can start think of how to transform it into software artefacts creating new UML diagrams where we dress up our concepts to form the functionality of our system.
Then you connect that functioality to your GUI designed by the methods you describe.
In this design process creating UML diagram not only makes communication easier, but it also makes the various steps in the design process more evident. This increases the chance that the right design decisions are taken by the right people and at the right time in the process.
If you only design with user inteface in mind chances are that your system will be hard to maintain in the long run.
Groups that all sound alike, and shaped in a form that people are supposed to like is all there is.
No thanks I'm not interested.
And even if there was something I liked I would think twice before buying. There are so much non standard, copy protected CDs around that may or may not be playable on my equippment, and I'm not taking the risk that they won't play.
Besides the fact that the CDs are copy protected, sort of implicitly accuses me of being a thief. I don't like that, or at least not trustworthy. So they can keep their music.
My guess is that it would take a week less time to have your developer up to speed with QT than what it would take if they used e.g. GTK. And there you have your $1000. Assume that a developer costs you $50/h and that he works 40h/week. That would give you a developer cost of $2000 in that week. This means that using QT have saved you $1000 in just one week. Your $1000 is probably less than what you pay for one sinle ad in a the press to get your application sold.
This line of reasoning is based on that QT actually is easier to use. Why do I think this is the case? Well, I was able to fix bugs and do minor application development after 3 hours in QT but in GTK I don't even know where to start. It could be that QT is better documented, but fact remains.
And if your application idea isn't unique enough so that it can't stand the competition from free sofware, then perhaps you should develop it as free software and and make your money on services related to it, or invent something better.
No not really. You can licence it through GPL, if you do open source work, or you can buy licences from Trolltech at a reasonalble price if you want to do closed source work.
The fact that you have to buy a licence for closed source may be a problem for hobby developers that have more time than money. To companies like Novell the quality is more important than the price . To them, time to market is everything. And they will spend far more money on developer saleries and marketing than they will spend on QT licences. So they will go for the best tool. And QT is one of the best there is. (Cross platform, good development tools, well designed framework...)
And if they want to release free software they have less risk of being hurt by some competitor if they release it under GPL than under LGPL or BSD type licences. I guess this was the reason MySQL AB changed the licence on their client libs to GPL.
So, QT would be a very good choise.
"security: OOo provides many of the same security features as MSO, but some of them are not needed for OOo"> Microsoft could actually be right when they claim that their Office suite have better security. This is because it supports TCPA. But to do that additional Microsoft software would be required. As this software runs on the server side it could also create a single point of failure, You would also almost certainly have to do a lot of eductation to both ordinary users and administrators. But security should not be something that resides at the application level. The OS should take care of such things. So if TCPA doesn't work with OOo we can only conclude that TCPA is flawed. Not that it is something wrong with OOo.