Not being a noob or anything like that, but I still wonder why should the operating system be case sensitive ? Any religious reason for that ? If you are talking about Linux's (if I could use that as an example for *nixes) usage in the mainstream, try explaining to an end-user that MyMail and mymail are different files.
I would disagree to that. Evaluating the merit of an update on the condition whether one needs to download the entire browser is unfair. Sometimes the APIs of the underlying libs change and that would mean relinking the dependant libs. In such a case, a simple overwrite might not work. I don't know the exact details, may be in this case it is required that you reinstall. Well, it is just a 5MB download !
But of course, one does need to give Microsoft the credit for being able to get the infrastructure for the security updates 'right'. May be they had few problems earlier on, but then, I think, they have set a precedent for such a mechanism.
Yeah, I think that definitely 'frees' Mono. Also if Mono implements the APIs provided by MS to ECMA then I think MS cannot do anything about it. Also I think there is more to Ximian than Mono. Lets not forget Evolution and the GNOME project. Evolution is a fine application for people moving to Linux. And the GNOME project is fast becoming the prefered desktop environment (and development environment too). So weighing Ximian (and Miguel in particular)only with respect to Mono would be baised. He has done a wonderful job with respect to OSS and I hope he continues to do the same in the future.
One thing that Mozilla has proved to the software community is that Open source projects can survive. And they can make a decent (if not in billions of $) profit. I think the main reason behind Mozilla's success are the developers. I am not talking about the on and off developers, but people like Asa Dotzler, Doug Turner, Christian Beisenger, Mitchell Baker, Boris Zbarsky, Brendan Eich, and the other drivers of Mozilla and not forgetting the countless others who are developing applications *using* Mozilla. Three cheers to everyone. I think the Mozilla foundation has evolved a wonderful business model for any entrepreneur looking to build software based on OSS.
Agreed, there is a thin line between the desktop and the browser. But look at Netscape's history itself. They dabbled with the idea of a comprehensive java solution which would make the browser - everything. The idea never took off. And for obvious reasons. There needs to be a clear demarcation between what a browser can and cannot do. You could go and add loads of features, but you need to remember that you as a browser developer donot own the operating system. Quirks in the OS can play a definite traunt.
And if Microsoft can do it with LongHorn, they can do it because they own the OS - they know how to use it the right way. I feel, it would be better if browser developers work more towards standards complaince (and in the case of firefox - mozilla) faster execution + smaller footprint.
That would definitely rake in the user count.
Definitely, the new look Wikipedia is wonderful to use.The latest news, the selected aniversaries and the did u know section were nice features thought by the folks there. Also the browse by section can be very handy. I have found Wikipedia's explaination on a wide range of topics very useful. It goes on to show how an open collaboration model can be made to produce wonderful results. And congratulations to the people at Wikipedia for achieving this landmark. I hope this prompts more people to contribute.
It is a common misconception that to acheive better performance, one needs to drill down to the assembly code. It has been shown that using VM languages like Java / C# people were able to get performances close to the compiled to native code languages. The trick is to know how the language behaves. A Simple thing like notifying the GC to delete an object can bring a good performance boost. I have also known applications that were written in C++ but with bad performance.
So I think to be able to get a good performance, knowledge of the language you are writing is very essential. I am not getting into the with such faster CPUs why do we have to worry about execution speeds argument
I do agree with OP on the question of the reviews of distros. Even the reviews are good in terms of the distro in particular, they lack the targetting of specific groups.
For example, something like, Shrike review for developers. Shrike does install the gtk libraries, but there *seems* to be no IDE like Anjuta. And if you want to listen to your favourite mp3s, well sorry folks, mp3 support is removed bcoz of the patent problems...yada yada yada.
Or something like Shrike review for grandma. If you want to listen to songs (the files with.mp3 extensions) then you will need to use the xmms player from Guru labs. And if you want to connect to the new digi-cam that you bought..well..yada yada..
I think reviews like these would be more helpful.
That was a lucid explaination, but what you are missing out is that there are so many implementations of the same thing. It is almost like that people have gone crazy about the specification from Microsoft (this is a rarity in the usually anti-Microsoft open source world). Would it not make more sense if the requirements and the specs are clearly marked out and both the parties decide whether (for eg.,) to do the drawing yourself (as in dotGNU) or rely on other libs(winelib or GTK#) for Mono.
I didnot know the existence of the dotGNU project till this post. Whereas Mono had been relatively well known. Going through the project page, the first reaction I had, was that there is no clear boundaries about where dotGNU ends and Mono starts (or otherwise). For example, both dotGNU and Mono have a C# compiler. Now how would a future contributor decide on which one to contribute for ?I think it would be very good for both the projects if they could have a simple explaination about what they are trying to accomplish and where (and how) they differ from each other.
Mono seems to be a very interesting way to go ahead, but what is more important is the fine print. Would Sun / IBM back Mono ? Would Sun want to nurture a competitor to its own framework (J2EE) ? And the interesting part is (as long as you donot worry about the proprietory APIs ) is that any.NET (mind you the one without any managed C++) app will run on Windows and Linux. This will work to Microsoft's advantage too - but the best thing for any developer would be the guarantee that his/her code will be cross platform. It would you the GDI calls on Windows and GTK calls on Linux ! Then you dont have to worry about the slow UI of Java.
These days more is heard of companies sueing each other, than the companies announcing something innovative. When was the last time that we have heard a really interesting and truly worth technology announcement from any of these big-wig companies ?
Please people, why do you forget, if you create something really worth its salt, you will get the customers automatically.
Agreed. The point is that the revenues that are gained by selling the software in turn goes to its employees and this in turns brings in more money into the market.
And as far as the innovation that the author of the article mentions, my only question for him is that, do all innovations need to manifest themselves directly to the end-user ? Not necessarily, is what I think. For eg., if a new scheduling algorithm is put in place, the user will only see a faster (say) load time of the app, but none in terms of features. So making a statement that there is no innovation at all is not being fair. I would have appreciated if the author gave more concrete proof.
kudos to real's employees for speaking out
on
Real's Reality
·
· Score: 2, Insightful
I donot know if Real managed to port their COM with XPCOM, but, that would help them leverage the functionality of a component object model for developers to exploit and would have helped them to gain acceptance from the open source community. There were more complaints about the install process and the capturing of the file types rather than the actual streaming technology itself. This is a shame for a technology company trying to provide an alternative for a/v streaming. May be the management needs to rethink its priorities regarding this.
And whoever the mailer(s) were, and if they are reading this post - nice to know that there are such honest people who are willing to voice their opinions.
I agree to your argument. But would the performance gain by dropping SVG pitch against the loss of a modular way to use the SVG renderer ? Assuming that there are xpcom interfaces for the SVG, then the SVG rendering engine (if it is not going to be tightly bound to Gecko) could be used by clients who might want (say) a charting engine based on xml. By dropping the xpcom interfaces to SVG arent the Moz developers loosing out on this ? Also SVG renderer might not be used very frequently. What is the point in trying to optimize something that is infrequently used ?
Agreed that any component object model (COM) is heavy and it does have its own problems. But the fact that Mozilla is built on a cross platform com is a huge advantage. If anyone wants to use these apis then they can do it without worrying about platform specifics. Even though currently xpcom is not very feature rich, it is a respected library. With everything else in the browser (or platform) running on xpcom, why do they specifically want to reduce the com support for SVG ?
Very true.The freedom of the open source world, lets individuals to take routes which might not work for the benefit of the project. In commercial software that is not the case. And that is one of the reasons why people prefer commercial software.
And as someone else pointed out, the post was more of a rant than a guideline. As someone who had a similar problem with printing with cups, I understand what ESR was trying to get at. Take Windows. Printing is a breeze in that. Even a techincal person doesnot have to wonder what is happening. Its just click click (and if unlucky get the install CD of Windows). I think these are somethings that can be imbibed for the general good of the project.
The sudden interest in Google because of the IPO might not work for the good of the company.Google is a great search engine.I cannot do without it. It embodies both the spirit of classy entrepreunership and technical excellence. To be able to continue the same the Google management needs to remain more techno-market savy rather than turn to being just market-savy. As the article points out, more often than not the companies that went public bore the brunt of the public intereference. Agreed that lots of revenues need to be raised for a good growth, but then it should be acknowledged that with great money comes great expectations and if Google cannot continue the same it would face the red line too.
Another thing to note in the excellence would be the lack of proper competition. If Teoma or Inktomi can deliver good results I am sure people would use that. This field is not a field of muscle but brains, and the smarter ones always win
There is a point in what Martin says. Just because there are a lot of eyes looking at the source it doesnot mean that it is secure. What is more important is that every piece of software undergoes a rigourous test procedure testing all (atleast most of) the possibilities.
And hitching on an anti-microsoft feeling would not help the open source world. If they want to prove that open source is better then they need to make sure to deliver quality products, and then the customers would adopt open source software.
Also take note of a valid point made by Martin when he quotes the diet coke example. Most of the linux distributions are overloaded with stuff that an average user would hardly use. It is not just the products but also the packaging that matters and if the linux distributions can pick up a cue from Microsoft I think there is nothing wrong in that. In the end it is the customer who will benefit.
Even though the security through obscurity has taken a beating, I think what the poster says is true. Even thpough the software may not be that secure, if the company doesnot derive any value from it, then they would not invest money to write ultra secure software.
You have to remember that its the 'managers' who make the decisions, and they look at roi, feasibility...blah blah...So the way out would be for you to write the ultra secure software and also write an exploit for the existing software and hope and pray the managers are in a good mood to listen to you !
good luck
Slashdot Mirror
Not being a noob or anything like that, but I still wonder why should the operating system be case sensitive ? Any religious reason for that ? If you are talking about Linux's (if I could use that as an example for *nixes) usage in the mainstream, try explaining to an end-user that MyMail and mymail are different files.
I would disagree to that. Evaluating the merit of an update on the condition whether one needs to download the entire browser is unfair. Sometimes the APIs of the underlying libs change and that would mean relinking the dependant libs. In such a case, a simple overwrite might not work. I don't know the exact details, may be in this case it is required that you reinstall. Well, it is just a 5MB download ! But of course, one does need to give Microsoft the credit for being able to get the infrastructure for the security updates 'right'. May be they had few problems earlier on, but then, I think, they have set a precedent for such a mechanism.
Yeah, I think that definitely 'frees' Mono. Also if Mono implements the APIs provided by MS to ECMA then I think MS cannot do anything about it. Also I think there is more to Ximian than Mono. Lets not forget Evolution and the GNOME project. Evolution is a fine application for people moving to Linux. And the GNOME project is fast becoming the prefered desktop environment (and development environment too). So weighing Ximian (and Miguel in particular)only with respect to Mono would be baised. He has done a wonderful job with respect to OSS and I hope he continues to do the same in the future.
One thing that Mozilla has proved to the software community is that Open source projects can survive. And they can make a decent (if not in billions of $) profit. I think the main reason behind Mozilla's success are the developers. I am not talking about the on and off developers, but people like Asa Dotzler, Doug Turner, Christian Beisenger, Mitchell Baker, Boris Zbarsky, Brendan Eich, and the other drivers of Mozilla and not forgetting the countless others who are developing applications *using* Mozilla. Three cheers to everyone. I think the Mozilla foundation has evolved a wonderful business model for any entrepreneur looking to build software based on OSS.
Agreed, there is a thin line between the desktop and the browser. But look at Netscape's history itself. They dabbled with the idea of a comprehensive java solution which would make the browser - everything. The idea never took off. And for obvious reasons. There needs to be a clear demarcation between what a browser can and cannot do. You could go and add loads of features, but you need to remember that you as a browser developer donot own the operating system. Quirks in the OS can play a definite traunt. And if Microsoft can do it with LongHorn, they can do it because they own the OS - they know how to use it the right way. I feel, it would be better if browser developers work more towards standards complaince (and in the case of firefox - mozilla) faster execution + smaller footprint. That would definitely rake in the user count.
Definitely, the new look Wikipedia is wonderful to use.The latest news, the selected aniversaries and the did u know section were nice features thought by the folks there. Also the browse by section can be very handy. I have found Wikipedia's explaination on a wide range of topics very useful. It goes on to show how an open collaboration model can be made to produce wonderful results. And congratulations to the people at Wikipedia for achieving this landmark. I hope this prompts more people to contribute.
It is a common misconception that to acheive better performance, one needs to drill down to the assembly code. It has been shown that using VM languages like Java / C# people were able to get performances close to the compiled to native code languages. The trick is to know how the language behaves. A Simple thing like notifying the GC to delete an object can bring a good performance boost. I have also known applications that were written in C++ but with bad performance. So I think to be able to get a good performance, knowledge of the language you are writing is very essential. I am not getting into the with such faster CPUs why do we have to worry about execution speeds argument
I do agree with OP on the question of the reviews of distros. Even the reviews are good in terms of the distro in particular, they lack the targetting of specific groups. For example, something like, Shrike review for developers. Shrike does install the gtk libraries, but there *seems* to be no IDE like Anjuta. And if you want to listen to your favourite mp3s, well sorry folks, mp3 support is removed bcoz of the patent problems...yada yada yada. Or something like Shrike review for grandma. If you want to listen to songs (the files with .mp3 extensions) then you will need to use the xmms player from Guru labs. And if you want to connect to the new digi-cam that you bought..well..yada yada..
I think reviews like these would be more helpful.
That was a lucid explaination, but what you are missing out is that there are so many implementations of the same thing. It is almost like that people have gone crazy about the specification from Microsoft (this is a rarity in the usually anti-Microsoft open source world). Would it not make more sense if the requirements and the specs are clearly marked out and both the parties decide whether (for eg.,) to do the drawing yourself (as in dotGNU) or rely on other libs(winelib or GTK#) for Mono.
I didnot know the existence of the dotGNU project till this post. Whereas Mono had been relatively well known. Going through the project page, the first reaction I had, was that there is no clear boundaries about where dotGNU ends and Mono starts (or otherwise). For example, both dotGNU and Mono have a C# compiler. Now how would a future contributor decide on which one to contribute for ?I think it would be very good for both the projects if they could have a simple explaination about what they are trying to accomplish and where (and how) they differ from each other.
Mono seems to be a very interesting way to go ahead, but what is more important is the fine print. Would Sun / IBM back Mono ? Would Sun want to nurture a competitor to its own framework (J2EE) ? And the interesting part is (as long as you donot worry about the proprietory APIs ) is that any .NET (mind you the one without any managed C++) app will run on Windows and Linux. This will work to Microsoft's advantage too - but the best thing for any developer would be the guarantee that his/her code will be cross platform. It would you the GDI calls on Windows and GTK calls on Linux ! Then you dont have to worry about the slow UI of Java.
These days more is heard of companies sueing each other, than the companies announcing something innovative. When was the last time that we have heard a really interesting and truly worth technology announcement from any of these big-wig companies ? Please people, why do you forget, if you create something really worth its salt, you will get the customers automatically.
Agreed. The point is that the revenues that are gained by selling the software in turn goes to its employees and this in turns brings in more money into the market. And as far as the innovation that the author of the article mentions, my only question for him is that, do all innovations need to manifest themselves directly to the end-user ? Not necessarily, is what I think. For eg., if a new scheduling algorithm is put in place, the user will only see a faster (say) load time of the app, but none in terms of features. So making a statement that there is no innovation at all is not being fair. I would have appreciated if the author gave more concrete proof.
I donot know if Real managed to port their COM with XPCOM, but, that would help them leverage the functionality of a component object model for developers to exploit and would have helped them to gain acceptance from the open source community. There were more complaints about the install process and the capturing of the file types rather than the actual streaming technology itself. This is a shame for a technology company trying to provide an alternative for a/v streaming. May be the management needs to rethink its priorities regarding this. And whoever the mailer(s) were, and if they are reading this post - nice to know that there are such honest people who are willing to voice their opinions.
I agree to your argument. But would the performance gain by dropping SVG pitch against the loss of a modular way to use the SVG renderer ? Assuming that there are xpcom interfaces for the SVG, then the SVG rendering engine (if it is not going to be tightly bound to Gecko) could be used by clients who might want (say) a charting engine based on xml. By dropping the xpcom interfaces to SVG arent the Moz developers loosing out on this ? Also SVG renderer might not be used very frequently. What is the point in trying to optimize something that is infrequently used ?
Agreed that any component object model (COM) is heavy and it does have its own problems. But the fact that Mozilla is built on a cross platform com is a huge advantage. If anyone wants to use these apis then they can do it without worrying about platform specifics. Even though currently xpcom is not very feature rich, it is a respected library. With everything else in the browser (or platform) running on xpcom, why do they specifically want to reduce the com support for SVG ?
Very true.The freedom of the open source world, lets individuals to take routes which might not work for the benefit of the project. In commercial software that is not the case. And that is one of the reasons why people prefer commercial software. And as someone else pointed out, the post was more of a rant than a guideline. As someone who had a similar problem with printing with cups, I understand what ESR was trying to get at. Take Windows. Printing is a breeze in that. Even a techincal person doesnot have to wonder what is happening. Its just click click (and if unlucky get the install CD of Windows). I think these are somethings that can be imbibed for the general good of the project.
Another thing to note in the excellence would be the lack of proper competition. If Teoma or Inktomi can deliver good results I am sure people would use that. This field is not a field of muscle but brains, and the smarter ones always win
There is a point in what Martin says. Just because there are a lot of eyes looking at the source it doesnot mean that it is secure. What is more important is that every piece of software undergoes a rigourous test procedure testing all (atleast most of) the possibilities. And hitching on an anti-microsoft feeling would not help the open source world. If they want to prove that open source is better then they need to make sure to deliver quality products, and then the customers would adopt open source software. Also take note of a valid point made by Martin when he quotes the diet coke example. Most of the linux distributions are overloaded with stuff that an average user would hardly use. It is not just the products but also the packaging that matters and if the linux distributions can pick up a cue from Microsoft I think there is nothing wrong in that. In the end it is the customer who will benefit.
Even though the security through obscurity has taken a beating, I think what the poster says is true. Even thpough the software may not be that secure, if the company doesnot derive any value from it, then they would not invest money to write ultra secure software. You have to remember that its the 'managers' who make the decisions, and they look at roi, feasibility...blah blah...So the way out would be for you to write the ultra secure software and also write an exploit for the existing software and hope and pray the managers are in a good mood to listen to you ! good luck