Re:Sean's speech at ESC about making a 3G devic
on
No More OpenMoko Phone
·
· Score: 2, Insightful
Uh, with the limited developer resources openmoko has always had, spending a year on a cancelled project is clearly poor management. Releasing the Freerunner with broken hardware is also poor management. I'd feel better about openmoko's chances as a company if they fired the management instead of all the developers. That never happens though - which is often why companies fail.
Makes me wonder how many good ideas are ruined by poor implementation. I'm betting this is a very large number. The problem is that people throw out the baby with the bathwater and so they might conclude that open-source phones are inherently a bad idea, instead of concluding that this group failed to design/produce them correctly.
Leonardo da Vinci sketched a car, and Karl benz according to the US patent office did more than any person to make it a reality, yet neither got the credit. Openmoko I think will be like that - computers are getting smaller, netbooks are the rage, projects like openPandora are pushing opensource hardware to its current limits, ie, the timing is right.
Its only a matter of time before decent management and funding will make an opensource phone happen. Maybe Android is it, but its just too much java imho to get the outside developer attention needed to make an embedded platform grow.
What I think is obvious to most of the people involved with openmoko is that they have had bad management, which will kill any project. They were always understaffed, and they spent a year on a canceled project. The fact that they pretend that their developers leaving is not a problem reminds me of several other experiences first hand - and none of those companies are here today.
Openmoko is now earily similair to a zombie company - keeps blowing sunshine while its developers quits or gets fired in droves, they stop building products, the only ones left are in marketing, and they linger on without doing much. The facts are:
1) The 10,000 phones are mostly of the 900mhz variety, which has a "buzz" issue that makes the phone unusable. You need to go to a "buzz fix" party to do a non-trivial hardware mod. The "A7" version that fixes these issues is in perpetual delay, with no release date in site.
2) The only two paid kernel developers have left this last month or have announced they are leaving, some key hardware guys have left in the last two months. Some key UI people have left over the last 6-8 months.
3) They've abandoned the next model, the GT03, and they have publicly stated no 3G without a guaranteed sales of 50,000 units.
I like the idea of Free software on mostly open hardware - only they can't for whatever reason get the hardware part right. I think the software is not the problem, its the hardware. The Freerunner has been described as a Porsche body with a lawnmower engine, and looking at openPandora, I scratch my head and wonder why its like that.
IMHO its like any project that is going down the tubes - far too few developers on a project changing scope too often.
Hardware's not easy - I damn near went insane from the politics of embedded linux projects myself - but I can't imagine working with a constantly changing hardware scope while everyone is leaving. I'd be pleasantly surprised if openmoko makes a comeback at this point - the first problem is I wonder how they could attract talent in the future, even if they could afford it.
"Just because a company has a huge and growing audience doesn't mean it can find a huge revenue source. Skype's appeal is that it offers services free or very cheap. That limits its ability to raise prices. And it turns out that there are limited opportunities for advertising or add-on services."
Skype-in and Skype-out are currently their main revenue sources, and they both have horrible quality. I live in Brazil and have family in the USA and other states in Brazil that do _not_ have internet. I tried skype-out with them and the quality was atrocious. I could have saved about 50% and its just not there yet.
In short, I've bought a skype phone that plugs directly in to my router since I exclusively run linux and their linux client sucks - I already happily spent US $250. I and my wife talks to my family that has internet all the time via skype to skype. Want more money from me? Fix skype-in and skype-out to have universally the same quality as a land line for 50% less money than the outrageous brazil telco's change, and I'll gladly pay you. I doubt they can and that's what I think it would take for skype to have a valid business model. Vonage etc isn't global so its wide open for skype themselves to win or lose.
I've also programmed in java for a long time - since '99. And I'm totally burnt out on the "yet another UI on top of a db" that seems par for the course in java - and in php, ruby, Qt etc for that matter so that's not really an exciting option for me
However, I programmed in C/C++ five years before java, and more importantly, I've been using linux since '96 and exclusively since '99. So I got an opportunity to port linux embedded to a custom ppc board, and yes, I'm doing PPC "Assembler on bare metal" to get the bootloader (u-boot) to initialize the cpu and memory so I can load the kernel. And I can say, once again, things are fun;-).
I'm on a project building a router using the freescale mpc8548e on a custom board - a pretty powerful processor for embedded work. This is my first embedded project, though I'm a long time linux user and had an early career doing primarily electronics work.
This book is uneven, but largely its a good book. The bootloader chapters are a very good intro - it even has a porting section going into the ppc assembly part of u-boot. The init sections of the kernel are also a good intro. The chapter describing gdb and kernel debugging is good, as I learned several tricks using macros with gdb that I think will help me a lot. Lots of tools like cpp are explained that I didn't know about.
However, where it comes short is that I still really need help with early debugging the bootloader and kernel init using the bdi-2000 (a hardware debugger that connects to the jtag interface that most boards have). It barely mentions the bdi-2000, which is a fundamental tool as its pretty much the only way to debug very early booting problems when porting.
I also think the flash section has too much of a jffs2 focus, and not ext2 focused to match most users. The driver section is very module focused and so bare it could be omitted.
Overall, I'm happy with the book in that it does have a lot of info that I haven't found elsewhere. It really convinced me I need to read the manuals of my hardware more as there is no alternative. The book has helped, and I think its a great intro.
I live in Brazil and I like to watch hockey. There is a package called "center ice" by the NHL, which allows you to stream games over the internet though its primrily a cable package. I'd pay to get some games - except I can't buy it because I live outside the usa and canada! So the only people that can pay to stream games, are people that can get them on TV. Even dumber, the games are blacked out if shown on another network - except you may be out of town which is one reason you'd stream games.
So I got my parents to setup a slingbox. I watched game two of the stanely cup last night - picture not bad really as it doesn't stall. Slingbox almost even runs on wine - it actually does for some people. So I'm happy and since these bigwigs don't care about me, I don't care about them as I didn't sign anything.
I call bullshit. I've travelled thru the rainforest twice, one for 3 weeks, and I didn't see any sugar cane at all. Lots of Mandioca is all. The brazilian rainforest simply a different type of soil not suited to sugar cane.
Sugar cane not only has a greater concentration of sucrose (about 30% more than corn) but it is also a lot easier to extract. Yet the USA places a 53 cent tarif on all imported ethanol. Powerful interests are at play, the greater good not being one of them. Brazil is lucky to be largely energy independant, which is in their politcal interest economically and security wise. The USA has double the oil of brazil with a roughly only a 30% larger population, but instead of being anywhere near energy independent, the USA imports 20% of its oil from Venezuela of which whose leader calls the USA president "the devil." Expect the USA to screw their corn industry, play brinkmanship with oil producing countries and thereby rising the price of oil, and continuing tarifs on importing ethanol. Confused? Follow the money and you may not be.
It means that any software that is developed within the next five years and is included with SUSE and might infringe on any patents filed by Microsoft up to the end of the agreement will be covered by the agreement.
This means that if software is written in 4 years, 300 days that is shipped with SUSE it would be covered forever for any Novell customers. Software that infringes on *new patents filed after the end of the agreement* would not be covered.'
Even if that is true, my three legal concerns are that the agreement as generally understood is revocable - a strange contract feature - and if microsoft disagrees with the interpretation above it can still drag things out in court at any time in the future by feat or by proxy. And lastly, I plan on using linux longer than 5 years and therefore this agreement leaves long term open questions. As an opensuse user I ironically feel far less safe.
You know, since novell has chosen to stand alone it may have to fight alone someday.
"Having the people who write and own the software turn upon Novell is a big deal to them. "
That's the ironic thing for me as a happy opensuse user, this deal has made me quite seriously thinking about switching distros for myself and my clients. Why?
1) I'm not comfortable with an agreement on linux that only supposedly applies to the novel. 2) Novel's answers implied that GPLv3 is unfinished, buts its clear to me GCC etc will become GPLv3 next year. That leaves me with a huge question mark for the future of opensuse. 3) The deal is revocable and in any event its only good for five years - an awfully strange contract IMHO. Both novell and microsoft are already disputing what it means - how is that supposed to make me feel good about a revocable contract good for only five years? 4) I can't remember any company benefiting from a microsoft agreement, which raises the question of their long term health. Having their revenue in the last few years be mostly from Microsoft deals doesn't help me feel better either.
In short, ever since novell bought Suse, every day its seems more like they are about to destroy it just like every other thing they have touched. Its a pity because although I consider opensuse to be a fine distro, I'm really thinking if I stay with them they will blow up in my face.
"Creating an American market for Brazillian ethanol likely encourages the destruction of that rainforest."
Maybe is just that. I'm an American expatriot living in Brazil, and last year I did a 20 day trip in the rain forest. I also own a flex fuel car. In short,
no one grows sugar cane in the rain forest. 99% of the people there do slash and burn farming for mandioca - a starch like a potato. Its just not the right land and weather for sugar cane.
Between fuel economy and price, ethanol in Brazil is about even. The huge pluses are that Brazil is essentially energy independant (1.7 million barrels of oil per day)in an age where the oil exporting countries are under threat to be invaded (Iran, venezuela), or are somewhat politcally unstable dictatorships with serious pressure groups (Saudi Arabia, Russia), You can always grow sugar cane, but some countries in the mid east such as Quatar are about to go dry. I consider fuel independance one of Brazil's biggest advantages in this century.
An over looked possibility is that USA crude-oil production capacity is about 5 million barrels per day and consumes 20 million barrels per day. A move to flex fuel cars in the USA could significantly reduce the price of oil - but I'd bet against it actually happening.
"And how will Trolltech still make money if they LGPL Qt ?"
Services and support, although admitedly that doesn't always work. At this point its a question of how much money they will lose if KDE becomes less popular. Furthermore, there are companies they make money with the LGPL - JBoss comes to mind.
Its Trolltechs decision of course - and they have changed their mind a few times. Recently they went GPL on win32, and of course their decision to go GPL in the first place.
One big question would be if more people would use KDE if it were LGPL - often people say "if only" but wouldn't change their mind anyways. I've seen this numerous times with Linux itself as it has overcome hurdles over the years.
My new sig - they'll take KDE away from me about the same time when they take away
my guns - they tried that recently here in Brasil and failed too.
iksrazal
One mans angst is another mans joy. Timing is everything, and Shuttleworth is cleverly showing some skill. I'm a long time Suse user and kubunto is now going on the next machine I have a chance to do a fresh install.
All Shuttleworth needs to do now is convince troll tech to move Qt to LGPL and I'll help build a shrine for him.
My new sig - They'll take KDE away from me about the same time when they take away
my guns - they tried that recently here in Brasil and failed too.
Not being LGPL is difficult for us who are independant app developers - I've actually written some Qt code. For Novell they have deep pockets that can make up for it in time saved, as they have done up until now. Qt tools and libs are pretty slick plus its C++ instead of C - better IMHO for desktop apps.
Of course Novell GPL'd Yast and other tools after they bought Suse, so for GPL apps it doesn't matter. Have any examples of non-GPL'd apps by Novell using gtk ?
My new sig:
- they'll take KDE away from me about the same time when they take away
my guns - they tried that recently here in Brasil and failed too.
I was afraid this would happen once novell bought Suse. To Novell's credit, up until now they have played smart - don't alienate your base users - primarily KDE users since pratically day one. Up until now what has Ximian and Evolution done for their bottom line? Mono? Puhlease. Suse Professional is their cash cow. Lose KDE and I lose Suse, I stop buying Suse professional, and I stop installing and recommending Suse to my clients who are spending top dollar - its that simple. I have my mom running Suse, my wife running Suse, my colleagues running Suse, and I install Suse for large telecoms. I lost redhat in 2002 after using it since 1996, and though I'd be sad for a while I'm sure I can switch again.
From what I have seen - unscientifically - KDE has been steadily gaining more market share then Gnome. I subscribe to the linux journals monthly desktop orientated pdf and they seem to agree. I have nothing against Gnome - I just happen to like KDE. Back in '99 I thought it was better for me and I have really liked KDE's progress ever since.
Where to go from here? First, I hope this is all wrong - I'm an enthusiastic Suse user. Kubuntu I suppose, but its a tough sell for my clients. Kooler heads prevail and I hope Novell is smarter than this, but somehow I doubt it. History shows Intel let the engineers create Itanium, and Novell has in the past bought Unix for top dollar and sell it to SCO for a huge loss, along with Corel etc.
I remember when Sun bought star office back in '99, they talked about a product called 'star portal' - a browser based office suite I think. Googling doesn't turn up much besides old press releases.
Given that history, all this could easily just be more vaporware.
"The US, of course, shouldn't be growing sugar at all at current prices; in a rational world its sugar industry would disappear and be replaced by imports from Brazil."
I'm American expatriot who lives in Brazil, whom also recently bought a hybrid VW Gol that runs on any combination of sugar-cane alcohol and gasoline. Nearly all cars sold in Brazil are hybrids.
Brazil has a unique advantage in that while its nearly self-dependant on oil, it also has a very strong agriculture sector. From a recent, no longer free nytime op-ed:
"During the 1973 Arab oil embargo Brazil was importing
almost 80 percent of its fuel supply," notes Mr. Luft,
director of the Institute for the Analysis of Global
Security. "Within three decades it cut its dependence
by more than half.... During that period the
Brazilians invested massively in a sugar-based ethanol
industry to the degree that about a third of the fuel
they use in their vehicles is domestically grown. They
also created a fleet that can accommodate this fuel."
Half the new cars sold this year in Brazil will run on
any combination of gasoline and ethanol. "Bringing
hydrocarbons and carbohydrates to live happily
together in the same fuel tank," he added, "has not
only made Brazil close to energy independence, but has
also insulated the Brazilian economy from the harming
impact of the current spike in oil prices."
"We are ready to import oil from Saudi Arabia but not
sugar from Brazil."
I think this article overlooks the fact that many 'free as in speech' third party security libraries and frameworks are available for java.
1) ACEGI - Aspect-orientaded-programming using a dependency injection model to replace or complement JAAS for authentication and authorization in an Application server independant way. A subproject of the Spring framework:
3) Container managed security implemented in every servlet container on the market, including tomcat.
In short, I'd like to see a comparison of the features and availablity of what people actually use in their applications, rather than an entirely fudgable comparison of reported/unreported security flaws.
"None are more hopelessly enslaved than those who falsely believe they
are free. -- Goethe"
"AOP in Java does smell a little like GOTO, IMHO. In brief, it uses "join points" to connect different aspects together (e.g. call this logging method just before calling that other method)."
I just don't get the GOTO reference that everyone talks about. Maybe in some narrow context but that would be largely missing the point. The Spring framework is probably the most popular AOP in Java, and while I'm just a beginner with Spring what I see is XML initialzing objects as beans.
Not proper XML in HTML/Slashdot but you get the point. So we have defined the myDAO interface being implemented by myDAOImpl - its pluggable. Then the setup is automatic in the code, done in a Service package similair to a Business Delegate in MVC:
public MyDAO getMyDAO() {
return MyDAO;
}
public void setMyDAO(MyDAO myDAO) {
this.myDAO = myDAO;
}
That and the replacement of JNDI getting DataSources and removing boiler plate code in things like JDBC and Hibernate. So to me, AOP reduces lines of code for common operations. What the hell does that have to do with GOTO?
Slashdot Mirror
Uh, with the limited developer resources openmoko has always had, spending a year on a cancelled project is clearly poor management. Releasing the Freerunner with broken hardware is also poor management. I'd feel better about openmoko's chances as a company if they fired the management instead of all the developers. That never happens though - which is often why companies fail.
Makes me wonder how many good ideas are ruined by poor implementation. I'm betting this is a very large number. The problem is that people throw out the baby with the bathwater and so they might conclude that open-source phones are inherently a bad idea, instead of concluding that this group failed to design/produce them correctly.
Leonardo da Vinci sketched a car, and Karl benz according to the US patent office did more than any person to make it a reality, yet neither got the credit. Openmoko I think will be like that - computers are getting smaller, netbooks are the rage, projects like openPandora are pushing opensource hardware to its current limits, ie, the timing is right.
Its only a matter of time before decent management and funding will make an opensource phone happen. Maybe Android is it, but its just too much java imho to get the outside developer attention needed to make an embedded platform grow.
What I think is obvious to most of the people involved with openmoko is that they have had bad management, which will kill any project. They were always understaffed, and they spent a year on a canceled project. The fact that they pretend that their developers leaving is not a problem reminds me of several other experiences first hand - and none of those companies are here today.
Openmoko is now earily similair to a zombie company - keeps blowing sunshine while its developers quits or gets fired in droves, they stop building products, the only ones left are in marketing, and they linger on without doing much. The facts are:
1) The 10,000 phones are mostly of the 900mhz variety, which has a "buzz" issue that makes the phone unusable. You need to go to a "buzz fix" party to do a non-trivial hardware mod. The "A7" version that fixes these issues is in perpetual delay, with no release date in site.
2) The only two paid kernel developers have left this last month or have announced they are leaving, some key hardware guys have left in the last two months. Some key UI people have left over the last 6-8 months.
3) They've abandoned the next model, the GT03, and they have publicly stated no 3G without a guaranteed sales of 50,000 units.
I like the idea of Free software on mostly open hardware - only they can't for whatever reason get the hardware part right. I think the software is not the problem, its the hardware. The Freerunner has been described as a Porsche body with a lawnmower engine, and looking at openPandora, I scratch my head and wonder why its like that.
IMHO its like any project that is going down the tubes - far too few developers on a project changing scope too often.
Hardware's not easy - I damn near went insane from the politics of embedded linux projects myself - but I can't imagine working with a constantly changing hardware scope while everyone is leaving. I'd be pleasantly surprised if openmoko makes a comeback at this point - the first problem is I wonder how they could attract talent in the future, even if they could afford it.
What would it take to port multics to my iPhone?
Kidding of course, I have ROKR and linux runs on it.
"Just because a company has a huge and growing audience doesn't mean it can find a huge revenue source. Skype's appeal is that it offers services free or very cheap. That limits its ability to raise prices. And it turns out that there are limited opportunities for advertising or add-on services."
Skype-in and Skype-out are currently their main revenue sources, and they both have horrible quality. I live in Brazil and have family in the USA and other states in Brazil that do _not_ have internet. I tried skype-out with them and the quality was atrocious. I could have saved about 50% and its just not there yet.
In short, I've bought a skype phone that plugs directly in to my router since I exclusively run linux and their linux client sucks - I already happily spent US $250. I and my wife talks to my family that has internet all the time via skype to skype. Want more money from me? Fix skype-in and skype-out to have universally the same quality as a land line for 50% less money than the outrageous brazil telco's change, and I'll gladly pay you. I doubt they can and that's what I think it would take for skype to have a valid business model. Vonage etc isn't global so its wide open for skype themselves to win or lose.
"I'm wondering where to go from here."
;-).
"Assembler on bare metal. It'll be a change."
I've also programmed in java for a long time - since '99. And I'm totally burnt out on the "yet another UI on top of a db" that seems par for the course in java - and in php, ruby, Qt etc for that matter so that's not really an exciting option for me
However, I programmed in C/C++ five years before java, and more importantly, I've been using linux since '96 and exclusively since '99. So I got an opportunity to port linux embedded to a custom ppc board, and yes, I'm doing PPC "Assembler on bare metal" to get the bootloader (u-boot) to initialize the cpu and memory so I can load the kernel. And I can say, once again, things are fun
I'm on a project building a router using the freescale mpc8548e on a custom board - a pretty powerful processor for embedded work. This is my first embedded project, though I'm a long time linux user and had an early career doing primarily electronics work.
This book is uneven, but largely its a good book. The bootloader chapters are a very good intro - it even has a porting section going into the ppc assembly part of u-boot. The init sections of the kernel are also a good intro. The chapter describing gdb and kernel debugging is good, as I learned several tricks using macros with gdb that I think will help me a lot. Lots of tools like cpp are explained that I didn't know about.
However, where it comes short is that I still really need help with early debugging the bootloader and kernel init using the bdi-2000 (a hardware debugger that connects to the jtag interface that most boards have). It barely mentions the bdi-2000, which is a fundamental tool as its pretty much the only way to debug very early booting problems when porting.
I also think the flash section has too much of a jffs2 focus, and not ext2 focused to match most users. The driver section is very module focused and so bare it could be omitted.
Overall, I'm happy with the book in that it does have a lot of info that I haven't found elsewhere. It really convinced me I need to read the manuals of my hardware more as there is no alternative. The book has helped, and I think its a great intro.
I live in Brazil and I like to watch hockey. There is a package called "center ice" by the NHL, which allows you to stream games over the internet though its primrily a cable package. I'd pay to get some games - except I can't buy it because I live outside the usa and canada! So the only people that can pay to stream games, are people that can get them on TV. Even dumber, the games are blacked out if shown on another network - except you may be out of town which is one reason you'd stream games.
So I got my parents to setup a slingbox. I watched game two of the stanely cup last night - picture not bad really as it doesn't stall. Slingbox almost even runs on wine - it actually does for some people. So I'm happy and since these bigwigs don't care about me, I don't care about them as I didn't sign anything.
Robert
I call bullshit. I've travelled thru the rainforest twice, one for 3 weeks, and I didn't see any sugar cane at all. Lots of Mandioca is all. The brazilian rainforest simply a different type of soil not suited to sugar cane.
Sugar cane not only has a greater concentration of sucrose (about 30% more than corn) but it is also a lot easier to extract. Yet the USA places a 53 cent tarif on all imported ethanol. Powerful interests are at play, the greater good not being one of them. Brazil is lucky to be largely energy independant, which is in their politcal interest economically and security wise. The USA has double the oil of brazil with a roughly only a 30% larger population, but instead of being anywhere near energy independent, the USA imports 20% of its oil from Venezuela of which whose leader calls the USA president "the devil." Expect the USA to screw their corn industry, play brinkmanship with oil producing countries and thereby rising the price of oil, and continuing tarifs on importing ethanol. Confused? Follow the money and you may not be.
You could buy a skype phone that runs linux internally and connects via dhcp directly to your router - no pc ;-)
http://www.sharperimage.com/us/en/catalog/product/ sku__ON509/
'The "5 year" is miss understood.
It means that any software that is developed within the next five years and is included with SUSE and might infringe on any patents filed by Microsoft up to the end of the agreement will be covered by the agreement.
This means that if software is written in 4 years, 300 days that is shipped with SUSE it would be covered forever for any Novell customers. Software that infringes on *new patents filed after the end of the agreement* would not be covered.'
Even if that is true, my three legal concerns are that the agreement as generally understood is revocable - a strange contract feature - and if microsoft disagrees with the interpretation above it can still drag things out in court at any time in the future by feat or by proxy. And lastly, I plan on using linux longer than 5 years and therefore this agreement leaves long term open questions. As an opensuse user I ironically feel far less safe.
You know, since novell has chosen to stand alone it may have to fight alone someday.
"Having the people who write and own the software turn upon Novell is a big deal to them. "
That's the ironic thing for me as a happy opensuse user, this deal has made me quite seriously thinking about switching distros for myself and my clients. Why?
1) I'm not comfortable with an agreement on linux that only supposedly applies to the novel.
2) Novel's answers implied that GPLv3 is unfinished, buts its clear to me GCC etc will become GPLv3 next year. That leaves me with a huge question mark for the future of opensuse.
3) The deal is revocable and in any event its only good for five years - an awfully strange contract IMHO. Both novell and microsoft are already disputing what it means - how is that supposed to make me feel good about a revocable contract good for only five years?
4) I can't remember any company benefiting from a microsoft agreement, which raises the question of their long term health. Having their revenue in the last few years be mostly from Microsoft deals doesn't help me feel better either.
In short, ever since novell bought Suse, every day its seems more like they are about to destroy it just like every other thing they have touched. Its a pity because although I consider opensuse to be a fine distro, I'm really thinking if I stay with them they will blow up in my face.
http://www.jguru.com/faq/view.jsp?EID=448031
iksrazal
Maybe is just that. I'm an American expatriot living in Brazil, and last year I did a 20 day trip in the rain forest. I also own a flex fuel car. In short, no one grows sugar cane in the rain forest. 99% of the people there do slash and burn farming for mandioca - a starch like a potato. Its just not the right land and weather for sugar cane.
Between fuel economy and price, ethanol in Brazil is about even. The huge pluses are that Brazil is essentially energy independant (1.7 million barrels of oil per day)in an age where the oil exporting countries are under threat to be invaded (Iran, venezuela), or are somewhat politcally unstable dictatorships with serious pressure groups (Saudi Arabia, Russia), You can always grow sugar cane, but some countries in the mid east such as Quatar are about to go dry. I consider fuel independance one of Brazil's biggest advantages in this century.
An over looked possibility is that USA crude-oil production capacity is about 5 million barrels per day and consumes 20 million barrels per day. A move to flex fuel cars in the USA could significantly reduce the price of oil - but I'd bet against it actually happening.
iksrazal
Services and support, although admitedly that doesn't always work. At this point its a question of how much money they will lose if KDE becomes less popular. Furthermore, there are companies they make money with the LGPL - JBoss comes to mind.
Its Trolltechs decision of course - and they have changed their mind a few times. Recently they went GPL on win32, and of course their decision to go GPL in the first place.
One big question would be if more people would use KDE if it were LGPL - often people say "if only" but wouldn't change their mind anyways. I've seen this numerous times with Linux itself as it has overcome hurdles over the years.
My new sig - they'll take KDE away from me about the same time when they take away my guns - they tried that recently here in Brasil and failed too.
iksrazal
All Shuttleworth needs to do now is convince troll tech to move Qt to LGPL and I'll help build a shrine for him.
My new sig - They'll take KDE away from me about the same time when they take away my guns - they tried that recently here in Brasil and failed too.
iksrazal
Of course Novell GPL'd Yast and other tools after they bought Suse, so for GPL apps it doesn't matter. Have any examples of non-GPL'd apps by Novell using gtk ?
My new sig: - they'll take KDE away from me about the same time when they take away my guns - they tried that recently here in Brasil and failed too.
iksrazal
From what I have seen - unscientifically - KDE has been steadily gaining more market share then Gnome. I subscribe to the linux journals monthly desktop orientated pdf and they seem to agree. I have nothing against Gnome - I just happen to like KDE. Back in '99 I thought it was better for me and I have really liked KDE's progress ever since.
Where to go from here? First, I hope this is all wrong - I'm an enthusiastic Suse user. Kubuntu I suppose, but its a tough sell for my clients. Kooler heads prevail and I hope Novell is smarter than this, but somehow I doubt it. History shows Intel let the engineers create Itanium, and Novell has in the past bought Unix for top dollar and sell it to SCO for a huge loss, along with Corel etc.
Say it aint so, novell.
iksrazal
Given that history, all this could easily just be more vaporware.
I'm American expatriot who lives in Brazil, whom also recently bought a hybrid VW Gol that runs on any combination of sugar-cane alcohol and gasoline. Nearly all cars sold in Brazil are hybrids.
Brazil has a unique advantage in that while its nearly self-dependant on oil, it also has a very strong agriculture sector. From a recent, no longer free nytime op-ed:
"During the 1973 Arab oil embargo Brazil was importing almost 80 percent of its fuel supply," notes Mr. Luft, director of the Institute for the Analysis of Global Security. "Within three decades it cut its dependence by more than half. ... During that period the
Brazilians invested massively in a sugar-based ethanol
industry to the degree that about a third of the fuel
they use in their vehicles is domestically grown. They
also created a fleet that can accommodate this fuel."
Half the new cars sold this year in Brazil will run on
any combination of gasoline and ethanol. "Bringing
hydrocarbons and carbohydrates to live happily
together in the same fuel tank," he added, "has not
only made Brazil close to energy independence, but has
also insulated the Brazilian economy from the harming
impact of the current spike in oil prices."
"We are ready to import oil from Saudi Arabia but not sugar from Brazil."
iksrazal
1) ACEGI - Aspect-orientaded-programming using a dependency injection model to replace or complement JAAS for authentication and authorization in an Application server independant way. A subproject of the Spring framework:
http://acegisecurity.sourceforge.net/docbook/acegi .html/
2) XML Encryption and XML Digital Signatures. Used in Web Service security or independently.
http://xml.apache.org/security/
http://ws.apache.org/wss4j/
3) Container managed security implemented in every servlet container on the market, including tomcat.
In short, I'd like to see a comparison of the features and availablity of what people actually use in their applications, rather than an entirely fudgable comparison of reported/unreported security flaws.
"None are more hopelessly enslaved than those who falsely believe they are free. -- Goethe"
iksrazal
I just don't get the GOTO reference that everyone talks about. Maybe in some narrow context but that would be largely missing the point. The Spring framework is probably the most popular AOP in Java, and while I'm just a beginner with Spring what I see is XML initialzing objects as beans.
bean id="myDAO" class="com.dao.myDAOImpl" property name="sessionFactory" ref bean="mySessionFactory"
Not proper XML in HTML/Slashdot but you get the point. So we have defined the myDAO interface being implemented by myDAOImpl - its pluggable. Then the setup is automatic in the code, done in a Service package similair to a Business Delegate in MVC:
public MyDAO getMyDAO() { return MyDAO; }
public void setMyDAO(MyDAO myDAO) { this.myDAO = myDAO; }
That and the replacement of JNDI getting DataSources and removing boiler plate code in things like JDBC and Hibernate. So to me, AOP reduces lines of code for common operations. What the hell does that have to do with GOTO?
iksrazal
Try this on a 16 gigs of memory Solaris 10K:
byte[] b = new byte[3000L * 1000L * 1000L];
Seen this question today:
http://groups.google.com.br/groups?hl=pt-BR&lr=&th readm=pan.2005.04.21.09.43.29.859952%40fffffffffff fff.com&rnum=11&prev=/groups%3Fq%3Dmemory%2B2005%2 6hl%3Dpt-BR%26lr%3D%26group%3Dcomp.lang.java.progr ammer%26start%3D10%26sa%3DN
People doing biometrics apparently are trying to.
iksrazal