Sunday, February 28, 2010
Saturday, February 20, 2010
Privacy
Privacy is an increasingly important and mistreated asset in today's digital world. With the likes of search data, purchase histories and many more information profiling (or gathering) systems are eroding very important and supposedly privacy that we imagine we have. This is certainly not the case as it is clear with very little research that there is significant shift in privacy. We pay for many free systems on the Internet with our very privacy that we have guarded for many years. It is very simple to find sex, birth date, address, friends etc. and this is enough information to track down a person, especially if search results are indexed and maintained.
This situation exists today and will grow significantly as more and more 'mash up' sites are created and gather more information from us with less security or badly programmed authentication systems.
maidsafe's perpetual data (pd) product presents a shift in this paradigm and allows people to access information freely and without tracking. In pd data is, well, just data. There is no access wall to data such as web servers, web browsers or traditional data access mechanisms that exist today. The data in maidsafe is 'on your local disk' or appears as 'physical data' on your hard drive on your computer or any computer you happen to log into and use. In pd
Pure data (pd) is a goal of maidsafe and unfettered access is paramount. The ability to simply open a file of interest or look at a photo or video of interest without divulging ANY information about yourself is key to the growth in data sharing and knowledge transfer.
This may not be seen as incredibly important today by the majority of people but it is appearing on the horizon as something perhaps to be aware of and will no doubt become one area of modern life we want to reclaim.
http://bit.ly/cxeH3C
for uk readers, interesting take on privacy. It's time for a new way, a network where privacy is primary ! It's time for perpetual data, it really is. http://ping.fm/HVXF4
Friday, February 19, 2010
Thursday, February 18, 2010
Tuesday, February 16, 2010
Monday, February 15, 2010
maidsafe-dht - c++ DHT (kademlia) with NAT traversal and cryptographic libraries http://ping.fm/OHI5y
Sunday, February 14, 2010
Perpetual Data overview
A sneak preview of what Perpetual Data does and looks like (for the moment anyway!). | Views: 168 0 ratings | |
Time: 02:57 | More in Science & Technology |
Startup hotshots are older and more educated than generally thought; it follows that training and finance skills should be offered to all age groups http://ping.fm/RA2K5
Friday, February 12, 2010
Tuesday, February 09, 2010
Revision 351: general tidy up :
Changed Paths:
Modify /trunk/src/CMakeLists.txt general tidy up http://bit.ly/bgKb7q
Changed Paths:
Modify /trunk/src/CMakeLists.txt general tidy up http://bit.ly/bgKb7q
Monday, February 08, 2010
Revision 350: Added TCP implemtation of Transport API and its tests.:
Changed Paths:
Modify /trunk/sr.. http://bit.ly/dcN5t7
Changed Paths:
Modify /trunk/sr.. http://bit.ly/dcN5t7
Sunday, February 07, 2010
Saturday, February 06, 2010
Why Open Source ?
It may appear like a strange move for maidsafe to release the foundation or core networking components of the system open source. In many ways with today's thinking that is true, but you have to look further to understand the reasons why Open Source in this case makes sense and will strengthen our product PD, which will not be open source (not initially).
What is the maidsafe-dht project ?
Basically it's a c++ implementation (lots of boost etc. easily adapted for API to other languages such as java, python, php etc.) which allows p2p or distributed system to be developed.
maidsafe-dht is a kademlia DHT implementation, with built in cryptographic libraries (courtesy of another open source project, cryptopp), a network layer (courtesy of another open source project udt) all wrapped in a nice clean API for people to connect with.
This allows a distributed hash table implementation of a network where you can easily store and find data of practically any type. The network has another API / modular approach and can allow udp, tcp etc. communications as well and importantly has built in NAT traversal methods such as UPnP and nat-pnp as well as a hole punching system very similar to ICE, but without the need for servers as all channels are made available via the dht itself.
This is a very valuable system and will allow p2p development to be carried out without huge up front costs for many people an businesses getting involved in p2p network.
Why give it away ?
The more appropriate question would be why do we think we can build test and maintain this part of the system better than hundreds or thousands of contributors and developers. The truth is we cannot and need to increase reliability, innovation in this area and frankly test it in many situations on many platforms (pc's mobile phones etc.).
It is like a car manufacturer giving away technical specs for a chassis, the car manufacturer may not sell or make money from the chassis, but does with what goes on the chassis. It makes sense all manufacturers use the same chassis for the same reason and they all work together to improve on it and allow small builders and individuals access to to the latest design advances and innovations in that arena.
So basically it simply makes a lot of sense and at no commercial cost.
Does that not encourage competition ?
Yes, and it should, the best products will succeed regardless. Competition is always a good thing and secrecy and not allowing access to information is perhaps not a great business model these days (thank goodness).
This is the first gift to the world maidsafe has made and already it is finding use in some companies, we expect a lot more soon and we know perpetual data will be a better product for it.
Will perpetual data ever be open source ?
In a word, yes and so it should. After launch we will identify revenue streams for maidsafe to continue to innovate. The focus of these revenue streams will be that which allow us to open source the product, although we may have to protect patents whilst doing so for a while (we hope not too long as if we stifle any innovation we will not be doing the right thing, and that's not acceptable for this project. We realise patents and open source in many ways is oil and water, but we also realise that there is always a correct answer to most problems or ssues such as this.
What revenue streams will exist in PD ?
We do not really know at this point, we may introduce a flat rate advertising model (which may be very different from traditional advertising), disk space selling, brokering disc sales between customers and many more. we do not want to pin any hopes on any particular stream just yet.
Can I help ?
Yes, sign up for the release (use anonymous mail if you can,we do not collect such data and we really do not want to know our customers, that way we cannot affect or harm your privacy).
We would love to hear comments and suggestions of how maidsafe should make revenues from such a network and do so as fairly as possible. This has to be done in a way that keeps everyone happy and satisfied we are doing something that is for the benefit of our vision of privacy, security and freedom for all. http://bit.ly/bpRHb9
What is the maidsafe-dht project ?
Basically it's a c++ implementation (lots of boost etc. easily adapted for API to other languages such as java, python, php etc.) which allows p2p or distributed system to be developed.
maidsafe-dht is a kademlia DHT implementation, with built in cryptographic libraries (courtesy of another open source project, cryptopp), a network layer (courtesy of another open source project udt) all wrapped in a nice clean API for people to connect with.
This allows a distributed hash table implementation of a network where you can easily store and find data of practically any type. The network has another API / modular approach and can allow udp, tcp etc. communications as well and importantly has built in NAT traversal methods such as UPnP and nat-pnp as well as a hole punching system very similar to ICE, but without the need for servers as all channels are made available via the dht itself.
This is a very valuable system and will allow p2p development to be carried out without huge up front costs for many people an businesses getting involved in p2p network.
Why give it away ?
The more appropriate question would be why do we think we can build test and maintain this part of the system better than hundreds or thousands of contributors and developers. The truth is we cannot and need to increase reliability, innovation in this area and frankly test it in many situations on many platforms (pc's mobile phones etc.).
It is like a car manufacturer giving away technical specs for a chassis, the car manufacturer may not sell or make money from the chassis, but does with what goes on the chassis. It makes sense all manufacturers use the same chassis for the same reason and they all work together to improve on it and allow small builders and individuals access to to the latest design advances and innovations in that arena.
So basically it simply makes a lot of sense and at no commercial cost.
Does that not encourage competition ?
Yes, and it should, the best products will succeed regardless. Competition is always a good thing and secrecy and not allowing access to information is perhaps not a great business model these days (thank goodness).
This is the first gift to the world maidsafe has made and already it is finding use in some companies, we expect a lot more soon and we know perpetual data will be a better product for it.
Will perpetual data ever be open source ?
In a word, yes and so it should. After launch we will identify revenue streams for maidsafe to continue to innovate. The focus of these revenue streams will be that which allow us to open source the product, although we may have to protect patents whilst doing so for a while (we hope not too long as if we stifle any innovation we will not be doing the right thing, and that's not acceptable for this project. We realise patents and open source in many ways is oil and water, but we also realise that there is always a correct answer to most problems or ssues such as this.
What revenue streams will exist in PD ?
We do not really know at this point, we may introduce a flat rate advertising model (which may be very different from traditional advertising), disk space selling, brokering disc sales between customers and many more. we do not want to pin any hopes on any particular stream just yet.
Can I help ?
Yes, sign up for the release (use anonymous mail if you can,we do not collect such data and we really do not want to know our customers, that way we cannot affect or harm your privacy).
We would love to hear comments and suggestions of how maidsafe should make revenues from such a network and do so as fairly as possible. This has to be done in a way that keeps everyone happy and satisfied we are doing something that is for the benefit of our vision of privacy, security and freedom for all. http://bit.ly/bpRHb9
Thursday, February 04, 2010
Revision 349: Adding CRYPTOPP_DISABLE_ASM flag to MacOS cryptopp cmake:
Changed Paths:
Modify /trunk/.. http://bit.ly/bLV1xR
Changed Paths:
Modify /trunk/.. http://bit.ly/bLV1xR
Wednesday, February 03, 2010
Revision 348: making the build a bit quieter - we have a few signed vs unsigned comparisons ne:
Changed Paths.. http://bit.ly/dxJeMa
Changed Paths.. http://bit.ly/dxJeMa
Revision 347: small changes back to almost clean udt now :
Changed Paths:
Modify /trunk/src/CMakeList.. http://bit.ly/aCQuKe
Changed Paths:
Modify /trunk/src/CMakeList.. http://bit.ly/aCQuKe
Tuesday, February 02, 2010
Revision 346: cryptopp_static proper now :
Changed Paths:
Modify /trunk/src/CMakeLists.txt Modify .. http://bit.ly/cxXCPy
Changed Paths:
Modify /trunk/src/CMakeLists.txt Modify .. http://bit.ly/cxXCPy
Monday, February 01, 2010
Revision 345: now compiles again release and debug - sorry folks :
Changed Paths:
Modify /trunk/src/C.. http://bit.ly/dn7A0N
Changed Paths:
Modify /trunk/src/C.. http://bit.ly/dn7A0N
Revision 344: fixed DLL_EXPORTS flag issue :
Changed Paths:
Modify /trunk/src/udt/CMakeLists.txt fixe.. http://bit.ly/9ft6gd
Changed Paths:
Modify /trunk/src/udt/CMakeLists.txt fixe.. http://bit.ly/9ft6gd
Revision 343: Added a properties flag to try and ensure we pick up correct libs when linking -:
Changed Paths.. http://bit.ly/a9q3Vf
Changed Paths.. http://bit.ly/a9q3Vf
Revision 341: OK commiting this to see if anyobody can spot the upnp error - it's a silly one :
Changed Paths.. http://bit.ly/8XQBbm
Changed Paths.. http://bit.ly/8XQBbm
Revision 339: added __WIN32__ definition in udt compile :
Changed Paths:
Modify /trunk/src/CMakeLists.. http://bit.ly/9xRh6O
Changed Paths:
Modify /trunk/src/CMakeLists.. http://bit.ly/9xRh6O
Subscribe to:
Posts (Atom)