If i had a software company i wouldn’t feel safe developing for Microsoft’s products, the wqay they treat their partners is not ethical (see groklaw.net and read a thing or two about the obscurity behind the corporation, the patent wars and anti-trust wars).

About the C issue: Every programmer in the world must know a little C, it’s the most basic and common software development tool and of course know well the fashion language.

And about .NET, there is no true legal support for developers to use .NET and about the ECMA standard is only c# and is not complete, several parts of .NET are obscurely patented and is not and it will not be open. And just for that .NET should not be considered an standard package for standard distributions, at least the people should know that the distro they are using contains non-standard packages.

No one knows when Microsoft will go against every distribution claiming they hold the rights for everything GNU/Linux had become thanks to .NET (i know it’s a very dramatic case but remember tom tom, and now TiVo, they waited some time and then attacked… i hope Attachmate Corporation be aware of this).

http://www.itwire.com/opinion-and-analysis/open-sauce/25215-the-elusive-royalty-free-patent-licence-for-mono

Open source is stagnating there is very little change – now some people who are new to Unix may find it all different ( or they came form MS world when the terrible windows ME ruled the roost) but when you started using Unix in the 80s and return to it 12 years later to find very little has changed its a bit disconcerting especially with the suite of MS latest offerings Windows 7 , Azure , WPF , WCF , ASP.NET ( MVC) , VS 2010 etc . (That said i think Microsoft are too afraid and risk adverse now they should have shipped Midori OS , their fold out tablet etc however their dev team has delivered great products) .

When im building my cross compilers and tryng to hack some libs to build im just going this makes DLL hell look like a picnic im spending forever in configuration . Compare to .NET i just drop the assembly on Mono linux or CE or 64 bit windows and it runs never a seg fault , never an error (except for assembly not found which is fixed by copying an assembly) . Java almost got there but not quite . Why cant Open source get closer to this !

Mono is one of the few examples where Open source has taken a good proprietry large program and made it almost feature complete open source ( Linux is another since Minix was proprietry) .

Like Linus said open source needs to use the input of companies more, even Microsoft , open source needs to stop looking at Microsoft and the past and start developing something new. How will software and operating systems look in 20 years time ? Or how can we improve the developer and user overall experience … rather than occasionally playing catch up . I guarantee while huge open source repositries for vi , emacs , awk , rm , se in c programs will still exist no one in their right mind will program a new application in it this is almost true today. C will not even be tought in Universities anymore outside some specialty courses ( just like assembler is now) .

Miguel has wasted a lot of time and effort, on something that no one really wants. Which is too bad, he’s one hell of a programmer, too bad he hasn’t been doing something productive.

This works ok for apps though facebooks while free , sells personal details and details to advertisers so they can place adds . Paying for a 300K-1M+ line lib or driver is a different matter your not going to get donations to even cover your dev costs and adds are difficult.. How to get money for a new KDE/Gnome replacement ? Maybe the lack of improvement in certain areas reflect the difficulty to comercialise / pay for the R&D – meanwhile things like axis get a steady stream of patches / updates since it suits the commercial interests.

Gnome and KDE have made improvements but in the scheme of things they are trivial and focused more on software quality then the product . Google chrome was commercial and is a good example they copied WPF ( which copied other people) and uses XML to render a GUI . Though it is vasty inferior to Silverlight/WPF its good enough for phones , Google bought the company and made it open source to suit its own commercial purposes ( stop Apple controlling hand helds ) – the original creators have been paid.

However that scheme is now obsoleted with a all new branch that are “online” projects. Right now, any famous online services has a FOSS or OSS project that can counter it (such as facebook, twitter, dropbox, youtube, gmail and many more)

All those project does not have any economic plan. They just live with donations and comunity help.

What is the meaning of those projects if it’s for being used by home users or one company ? It has it’s use but very limited ….. and going further it’s impossible due to licenses witch are too much restrictive.

When it’s time to go to the “mass use” and Commercial use you confront different situations :
- Like Mysql AB they “pervert” the spirit and you sale at the same price than proprietary software (or even more)
- Like Google you use it, you modify it, you massively gain money with it but you don’t share. It’s the famous “know how” that one proprietary company will protect that make his product unique
- Like Sun you make a bait (openjdk) so that the OSS/FOSS comunity will not complain about having FOSS from java and blur/evangelize about their technology that, in fact, is totaly proprietary.
- Like so many companies that have tried to promote OSS and ends and dies in trials with the FSF or equivalent because they have not respected 100% of the licence. (witch is normal to be punished for that but make also collateral damage to OSS/FOSS and that may be even greater than not respecting the license)

It’s a loose/loose scheme in 90% of the cases. And in front of that there is just a static positions from the “big institutions” of the Linux ecosystem.

The battle theme :
battle can be healthy when they are won …. and winning is consisting of gaining the other camp territory, not having more territory. If it’s for having a battle that is a never ending battle it’s not healthy and more when there is others in the play (google and other projects that evolves quicker). When there is a common threat the common sense is to stop the war and to stick together.
Let me show you how bad this is for KDE and GNOME right now :
1 – KDE and GNOME are making battle one against another.
2 – KDE has made a big mistake with the 4th version and Gnome is about to make the same error with the 3rd version
3 – Inside GNOME there is many wars just like this one : having or not Mono and Mono apps
4 – KDE has a very poor app support due to their error with the 4th version, the kde apps are buggy and incomplete mostly
5 – Google announce an OS based on linux but without KDE or GNOME ….. with their own made window manager
6 – Reaction of KDE and GNOME ? still making war between each other without learning from each other mistake. And we all know how well done Google projects are.
7 – Possible result in the future : disparition of KDE and GNOME in the Linux ecosystem.

Competition is ok, battle are not in many times. The difference between the two is that in one you react and you are creative, in the other you just try to find how to attack your opponent.

I’m bored to read how much KDE and GNOME have changed since XX version ….. in the fact they are exactly the same as 10 years ago with just higher resolutions and better image presence or with better tweak support. They just find a way to do the same or what already exist with different techniques. And this applies to the other existing windows manager …… once created they really don’t change a lot.

On .NET in windows you downlaod the app and it runs it doesnt matter if its 32 bit , 64 bit or Arm.

“There is a lot more dev people that make programs for Linux than for Microsoft. It is said and known that Linux has the biggest dev population in the world. However Linux can’t even begin to take over Windows : Why is that ?”

Im not sure this is true – there are lots of part time / hobby Linux devs ( compared to MS where is 95% 9-5 devs)

“1 – The user habits ….. it’s complicated to change habits. (Mono is one of many response to that problem).”

Agreed . Users want somethin like OSX / Ipod . Remove the 1970s command line /shell crutch and instead of being an OS system admins loves it will become one users love. ( note im old and from the 70 ‘s )

“2 – Too much internal fights : KDE vs GNOME, Debian vs Ubuntu, emacs vs vi, GNU vs Linux kernel, and here : Mono vs some users (sorry, it’s difficult to define the group here : it’s not GNU, it’s not against microsoft ….)”

The fights are good.. eg Java vs .NET ( mono) , OSX vs WIndows and even KDE vs GNOME , WCF vs axis etc its survival of the fittest and all of these have led to improved products . What is not good is the lack of standards ( and dont speak to me of comitees like WS) eg OSX allow you to have Java , C++ , mono apps etc but they define a very consistant UI for app developers to follow. Without this consistancy devs just do their own thig which creates a bad user experience ( Even Linux command lines options are not consistant ) . Only someone like Linus could have the clout to say for a runs on Linux logo you must follow this but its against the anarchistic development we have now which attracts hobyists.

”
3 – The meritocracy system that has shown it’s Limits : in the last 5 years I haven’t seen any improvement of any OSS o FOSS system. The only thing that I’ve seen is an increase of egocentric moves like : “you can use my software as you please but you can’t sell it or having commercial use of it” or like when I read recently “let’s call Linux as GNU/Linux for giving merit to GNU” or simply when you have someone like Google that comes in by surprise and make android or chrome and nothing is made to counter it ….. (the worse is that there is projects for counter them …… but there is no moves to support them massively …. many collects critics instead just like Mono right now).
”
I returned to Linux 5 years ago ( i was involved in UNix in the 80s , Solaris in the late 90) and was astounded by the lack of change/progress and how small teams were though the Gnu c compiler actually seems to be low in bugs the last few years it wasnt always so

“you can use my software as you please but you can’t sell it or having commercial use of it” .

I dont think this is ego centric , I have worked with 3 open source projects and the model is rarely successfull. When its successfull there tends to be a single key driver ( either person or Company) who does the bulk and when almost finished or there is a product people join . However because of this creating a large initial code base from scratch , like something to challenge Flash , WPF /Silverlight is very difficult and it is easier done by a company. Its ironic that i dont think Linux would survive now if it started as a OS project the fact that Linus put his soul into it with his initial 10K lines of code but creating it now is no longer an option.

Personally im working on a new OS ( see http://www.shanghai-software.com/blog) based on ucLinux and borrowing some ideas from Android , Coyotos and Singularity. It will be more stable , more reliable , more secure and will run at the same speed . Now while im an experienced dev with OS experience this is a 300K line project ( at least) and i will hire people to build this out of my own pocket ( in China where i have moved to save costs and im safe from lawyers) and hope to sell to the embedded and security markets so will offer a non commercial licence. I have tried to attract devs but these tend to be kids who think its cool , know very little and cost more time to teach than they produce . Very few devs with 10+ years experience work in new open source projects in their own time.

Obviously if the projects fails it will change to a BSD licence but after all since im putting up my dough and time on a product that might help people ( porting apps in embedded systems is a royal and expensive pain) and hence I should be allowed a non commercial licence , after all why should people make money of my work / contribution there is no other field in the world like that eg entertainers , sports people etc yet devs often contribute more . I think Linus has some good comments on how companies should be part of open source.

Two things are very irritating :

1 – The speed of mono releases is very fast where it’s not the case for .net releases in the Microsoft ecosystem.
The result is that in the Microsoft world there is not a lot of trouble to know what you need to make your app run …… this case is made complicated with so much versions of mono present in the Linux World.
Other side effect of “speed vs reason” is that we have a preview of moonlight 3 since a lot of time ……. nothing real since the last winter Olympics ……. The project is stuck ? they have done a bad implementation that need a complete rewrite ? who knows.

2 – You may laugh with this one : the lack of real support from the Ubuntu/Debian Package maintainers ….. They stick with the 2.4 branch where this branch is about to be depreciated and has it’s stable replacement since more than a year. since 3 ubuntu release they have been ignoring 2.6 branch. Worse : they always stick to the past release of 2.4 branch …… (eg if the 2.4.32 was available they would use instead the 2.4.24). Why I complain about that ? It’s simple : you do it or you don’t do it but you must not stay between those two. By making those severe decisions the Ubuntu/Debian distros are one of the oldest distro in the Mono point of view …… witch make life to developers very complex (preview his app to run in Ubuntu/Debian and the others).

The magical thing in the Linux World is that, like this very case in Ubuntu, Microsoft doesn’t even need to try to make fuzz over Linux …… Due to the “misunderstood” concept of freedom there is a lot of fight between Linux groups :

There is a lot more dev people that make programs for Linux than for Microsoft. It is said and known that Linux has the biggest dev population in the world. However Linux can’t even begin to take over Windows : Why is that ?
1 – The user habits ….. it’s complicated to change habits. (Mono is one of many response to that problem).
2 – Too much internal fights : KDE vs GNOME, Debian vs Ubuntu, emacs vs vi, GNU vs Linux kernel, and here : Mono vs some users (sorry, it’s difficult to define the group here : it’s not GNU, it’s not against microsoft ….)
3 – The meritocracy system that has shown it’s Limits : in the last 5 years I haven’t seen any improvement of any OSS o FOSS system. The only thing that I’ve seen is an increase of egocentric moves like : “you can use my software as you please but you can’t sell it or having commercial use of it” or like when I read recently “let’s call Linux as GNU/Linux for giving merit to GNU” or simply when you have someone like Google that comes in by surprise and make android or chrome and nothing is made to counter it ….. (the worse is that there is projects for counter them …… but there is no moves to support them massively …. many collects critics instead just like Mono right now).

Divided it’s a lost battle.

That was just a little yell about it

Well depends on the distro – For a complete distor like Ubuntu which supports running out of the box it should be part of it . For a minalist distro remove it allong with everything uncessary like libc , kernel headers and a compiler.

Im beginning to really dislike the package manager mess with versioning issues ( makes DLL hell in windows look tame) and think bloatware OS but tested to work is looking good, Frustrated trying to get an app to work the last few days .

Removing or adding something by yourself is freedom (because it’s your decision)

to have it installed or not in the mainstream is not freedom (because someone made the decision instead of you)

So here we are in the “what is the first : the egg or the chicken ?” paradox : Must we let mono installed by default and simplify his removal or Must we remove mono from main branch and simplify his installation ?

In both case : we quit and give freedom to users

When we think “must we remove or add this or that on mainstream” it’s a freedom nonsense because the simple fact to choose one or another means quit freedom to part of users.

Hope that helps to understand what freedom is and is not.

“One persons freedom ends where another persons freedom begins” ….. if it’s not the case it’s not freedom: it’s called anarchy (everyone is free to do what he wants).