Jari development has a new targets in development. The first of all we're decided to move our dev/hours to the exiting GNU mach microkernel and implement our ideas based on it. This decision based on the real time needs via the following points :
- there are many trivial task thats simply takes the time on its
- there are already exiting code base that can be used for Jari ideas
- there are no time to create a newly microkernel from scratch
The possible ways may be different and depends on the development flow in the time.
I don't know yet if we will takes the GNU Hurd servers to modify and extend, but you can be sure we will split Jari ikernel and Jari EZA code base with the GNU Mach code base.
With this objectivies I hope that we will ran in the early dates.
Now, I've planned to get Jari run on the third decade of the 2009 with the minimal drivers set, libc, CLI and varios utilities.
Look out this blog for further details in the future.
Monday, October 01, 2007
Wednesday, April 11, 2007
Jari: small new things
While the low level under the heavy development stage, I'm returned to the compiler and some software. But about it by the way ...
C compiler:
We're developing the new C dialect, eC - stands for extended C.
No, it doesn't mean the next C++ or something else. Our dialect targeted to functional programming. I will extend the C macros, add the lambda functionality, and optimize all works with the types.
Also eC compiler will have a mechanism that will checks the works with pointers and variables to avoid the double malloc/free, it will be like an option.
Backing to C, it will backward compatible i.e. you will be able to compile C sources with eC compiler.
About this and other new features I will blog separately some later.
LibC/LibeC:
We will include the several memory allocation models, we will provide the standart alloc() and free(), but with it we're can offer the wrappers and callbacks to make the internal memory allocator simply and faster. You will be able to create struct mem_alloc_t and use it with your own way, with the libc/libec standart functions/politics or assign your function that preffered for your application. Also the libs will have several new functions for allocating memory - rt_alloc(), local_alloc(), ext_alloc(), func_alloc(), there functions will provide the more flexible parallel programming issues.
In addition, libs will be extended for string operating functions, the libec will have the standart high customizable parser.
coreutils:
Core utils will includes tools for parallel task management, via this tools you can get ran processes via your cluster by the your custom way, run nonlocal architecture binaries on the needed node and other features.
See for next blog messages, I will give here more examples and will describe all things written above.
C compiler:
We're developing the new C dialect, eC - stands for extended C.
No, it doesn't mean the next C++ or something else. Our dialect targeted to functional programming. I will extend the C macros, add the lambda functionality, and optimize all works with the types.
Also eC compiler will have a mechanism that will checks the works with pointers and variables to avoid the double malloc/free, it will be like an option.
Backing to C, it will backward compatible i.e. you will be able to compile C sources with eC compiler.
About this and other new features I will blog separately some later.
LibC/LibeC:
We will include the several memory allocation models, we will provide the standart alloc() and free(), but with it we're can offer the wrappers and callbacks to make the internal memory allocator simply and faster. You will be able to create struct mem_alloc_t and use it with your own way, with the libc/libec standart functions/politics or assign your function that preffered for your application. Also the libs will have several new functions for allocating memory - rt_alloc(), local_alloc(), ext_alloc(), func_alloc(), there functions will provide the more flexible parallel programming issues.
In addition, libs will be extended for string operating functions, the libec will have the standart high customizable parser.
coreutils:
Core utils will includes tools for parallel task management, via this tools you can get ran processes via your cluster by the your custom way, run nonlocal architecture binaries on the needed node and other features.
See for next blog messages, I will give here more examples and will describe all things written above.
Wednesday, January 24, 2007
offtopic: m$ the big technologies brake
Microsoft is a big shit deals company, but it's sounds like fanatic scream.
Okay, take a look for the arguments.
The first of all and generally people says - "hey, solution from microsoft are totally used and it tells that it's a better solutions" - it's a typical disunderstanding about technology and marketing targeted *only* for domination and "big money must came to us totally". We're can compare non-IT products, but products that really depends on the technologies, in example - every day used thing A. What are you doing before the "big bought" ? You compare the avialable things on the market, what are principles of your choose? Generally you looking for a better look&feel, after this you are look for the functionality, price, quality etc ... Are you sure on this? I don't sure, the first you spamed via the advertisment and marketing action like - "hey look at this, it's a used by everyone, it's better ..." and for more than 60 percent this thing will be preffered on your choose.
The same deal are applicable for the IT market, the end-user (non proffesional) will choose the product that has a more advertisment and if advertisment and marketing will be more aggressive the customer will be more targeted to the product. So, imaginate that a millions end-users attacked via the aggressive marketing, theirs don't thinking about functionality, there are thinking "if an advertisment is so sounds good, then product is good too". It's not a problem of the end-users it's a just big lack for bad business. Microsoft uses this way, it's mean that they don't improve the new features, stability, innovation and via intercepting the whole market they are kill all the ways to get the another concurrents, following this way the new thing in the business came - "Why we should to do something new and better if our products are buying ?". There are no concurrents and if there are coming to the market the microsoft killing its via the agressive marketing and brow up the set of the myth like "in the foo OS no stable, no features, it's a bad - choose our ones", or following with the another way - just buying another companies and closing the projects (remember Xenix for example).
But if you can see there are no fundamentally changes on the microsoft products long the many years. New GUI? New incompatible within formats? New animations? New installation programs? it's a things just for emulating that they doing something - for customers, otherwise customers didn't buy the new products and didn't pay for the microsoft.
But there are some limitations, the first leak of this business strategy that world contains not only end-users, the second leak is that selling products an ugly, the third leak that anyway microsoft must support their products, in short words I mean that there are many professional people that can analyze with system methods exiting solution and tell - "hey! folks! it's a big piece of shit!" and they are will be right. The other thing it's a critical industry that includes an industry automatic systems, science research that needs no for good GUI or pretty looks pooks, they are need for stable and featured product, that has a functionality, it's a 24x7 servers, embedded systems, industrial system, computer clusters, distributed networks, and other critical systems.
In this case microsoft will never shift out other, better solution from this segments. But thinks, that segments has a new innovations and for working with them end-user need have a features and new functionality, but microsoft doesn't do it or do it too slowly and ugly (remember when microsoft approved the tcp/ip stack and improve network support for their products and how it's ugly implemented for the current time). Microsoft sets a technology brake for others including developers, end-users, science, etc ...
But it cannot be eternal process, and now microsoft marketing department know about it, but there doesn't follow via the right way - the right way is a perform a better solutions in a products, compare win 200 with win xp - do you something new, excluding GUI look ? There are nothing, take a look for vista_ something new, excluding poor resource using and new GUI ? The answer is no, so why the microsoft will lay to the customers that linux is bad, that linux is old, that price of the linux solution is big - okay okay, but what the things are microsoft can offer for exchanging unixes ? There are nothing.
Think deeper, microsoft will die or redirected to the support and porting their products to the linux and get this segment.
Questions?
Okay, take a look for the arguments.
The first of all and generally people says - "hey, solution from microsoft are totally used and it tells that it's a better solutions" - it's a typical disunderstanding about technology and marketing targeted *only* for domination and "big money must came to us totally". We're can compare non-IT products, but products that really depends on the technologies, in example - every day used thing A. What are you doing before the "big bought" ? You compare the avialable things on the market, what are principles of your choose? Generally you looking for a better look&feel, after this you are look for the functionality, price, quality etc ... Are you sure on this? I don't sure, the first you spamed via the advertisment and marketing action like - "hey look at this, it's a used by everyone, it's better ..." and for more than 60 percent this thing will be preffered on your choose.
The same deal are applicable for the IT market, the end-user (non proffesional) will choose the product that has a more advertisment and if advertisment and marketing will be more aggressive the customer will be more targeted to the product. So, imaginate that a millions end-users attacked via the aggressive marketing, theirs don't thinking about functionality, there are thinking "if an advertisment is so sounds good, then product is good too". It's not a problem of the end-users it's a just big lack for bad business. Microsoft uses this way, it's mean that they don't improve the new features, stability, innovation and via intercepting the whole market they are kill all the ways to get the another concurrents, following this way the new thing in the business came - "Why we should to do something new and better if our products are buying ?". There are no concurrents and if there are coming to the market the microsoft killing its via the agressive marketing and brow up the set of the myth like "in the foo OS no stable, no features, it's a bad - choose our ones", or following with the another way - just buying another companies and closing the projects (remember Xenix for example).
But if you can see there are no fundamentally changes on the microsoft products long the many years. New GUI? New incompatible within formats? New animations? New installation programs? it's a things just for emulating that they doing something - for customers, otherwise customers didn't buy the new products and didn't pay for the microsoft.
But there are some limitations, the first leak of this business strategy that world contains not only end-users, the second leak is that selling products an ugly, the third leak that anyway microsoft must support their products, in short words I mean that there are many professional people that can analyze with system methods exiting solution and tell - "hey! folks! it's a big piece of shit!" and they are will be right. The other thing it's a critical industry that includes an industry automatic systems, science research that needs no for good GUI or pretty looks pooks, they are need for stable and featured product, that has a functionality, it's a 24x7 servers, embedded systems, industrial system, computer clusters, distributed networks, and other critical systems.
In this case microsoft will never shift out other, better solution from this segments. But thinks, that segments has a new innovations and for working with them end-user need have a features and new functionality, but microsoft doesn't do it or do it too slowly and ugly (remember when microsoft approved the tcp/ip stack and improve network support for their products and how it's ugly implemented for the current time). Microsoft sets a technology brake for others including developers, end-users, science, etc ...
But it cannot be eternal process, and now microsoft marketing department know about it, but there doesn't follow via the right way - the right way is a perform a better solutions in a products, compare win 200 with win xp - do you something new, excluding GUI look ? There are nothing, take a look for vista_ something new, excluding poor resource using and new GUI ? The answer is no, so why the microsoft will lay to the customers that linux is bad, that linux is old, that price of the linux solution is big - okay okay, but what the things are microsoft can offer for exchanging unixes ? There are nothing.
Think deeper, microsoft will die or redirected to the support and porting their products to the linux and get this segment.
Questions?
Subscribe to:
Posts (Atom)
