This will be the second post in the series of chronicling the events that happened in the second HamaraLinux which was held at Courtyard Mariott on 20th September 2015. The talks and sharing were given by Ameya, Nikhil and Gurvinder on various topics.
Going on from the last report, after Krishnakant’s talk, I had invited Ameya. Ameya was one of the fortunate ones who got the chance to attend and be part of both the Debcamp and Debconf held just the month before at Germany 2015. I was looking forward to seeing Ameya sharing about his experiences at Debconf. After a bit of egging, he did share about his experience of Debconf. I was expecting him to add a bit of pizzaz or something as most of the Gods, or at least most we think of as gods were there.
For reasons unknown, he got nervous and shared a somewhat superficial sharing about the Debconf and ended it in 10-15 mins. I was expecting him to share about what workshops if any he attended, met who’s who and what sort of roadmap did the various teams were working on internally. Setting aside the beer, the drinks, the cheese and various types of food which you will see in most of debconf reports, which celebrates the diversity and richness of the project, the main reason though that people go to Debcamp and Debconf is people want to get together to share where they want to go ahead and to have feedback from users, other developers etc.
Face-time is expensive but at the same time it is extremely important as friendships are made, knowledge and skills are exchanged, bug-reports are resolved, mis-understandings (if any) are cleared up and generally motivates people to give and be better than they are/were before . Debcamp is more for those who want to contribute by coding and for new and old programmers to get their feet on Debian Infrastructure while Debconf is more for networking, socializing, some coding and making grand plans to be the Universal Operating for everyone 🙂 .
Anyways, after Ameya’s sharing, it was now turn of Nikhil’s turn to share. Nikhil has been also a friend for quite some time, we actually met each other way back at some FOSS conference few years ago. Nikhil has been particularly trying to get the bus system in Pune fixed right for quite sometime now. The idea is simple, if we want to have a digital system where you can have all the bus stops in multi-lingual state, i.e. English, Hindi, Marathi it needs to be in a foss format, something which is open as obviously as Pune itself is growing, Buses which are environmentally friendly mode of transport and can haul quite a few people in one direction or the other. They are cheaper to run than Railways or Metro to run and doesn’t need dedicated infrastructure for the same (except for a shed to put them at night and do maintenance on them.) but then that is common for all and any means of public transport . So he’s working on first getting that data streamlined and then overlaying that data on Openstreetmap at some point in time and if those places are not mapped then either doing some quick mapping of the place or getting it done by interested people. In the long-term it is the people who will benefit.
It is actually a long-term project and he has been trying to get some students interested in the project. Nobody wants to fund the initial part as PMC coffers are empty and all would be happy to fund the implementation as there would be quick returns if it becomes an app. which it eventually will at some point of time. As I knew the project and do have an emotional connect as I used to take buses till the end of college life. So do understand the need of millions of masses and know that private vehicles is not the answer. After Nikhil’s sharing of the project, we broke for lunch.
After lunch it was now turn of Gurvinder to spell-bind us with his presentation about the long-range wireless project. Gurvinder started with sharing about different early modes of communication and there were also some prizes that were won by people ( Hamara T-shirts and pen drives) . After that Gurvinder went into explanations and differences between CISC and RISC. As of now, all ARM and Power v8 IBM processors are RISC based chips while all AMD and Intel are CISC-based chips. The main differences are differences in pre-programmed and thermal balance and limits. For e.g. both ARM and V8 would probably be a good choice for some embedded single specialized purpose project or task, while you need CISC chips for general purpose computing where it is a multi-purpose chip.
I should have asked Gurvinder to share the flags which would also have cleared lot of doubts :-
$ cat /proc/cpuinfo | grep flags
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm ida arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt
Now this is from my lappy and actually when I run the command, I get it 4 times indicating/telling me that I own a quad-core processor. I will not go into much details of the flags and simply share one of the links on wikipedia with one of the well-known flags shown above.
Similarly, you could just look up all the flag acronyms on wikipedia and you should get an idea of all the instructions that are pre-programmed in this chip or any other CISC chip that you have and are using under GNU/Linux. What not many people know that Intel has been quietly also investing in RISC chips and research from last few years so that they can hedge their bets in case of breakthoughs which make RISC a winner. In some ways it already has, as smartphones already outrank all the laptops and desktops that are sold in the world. On top of that they are growing at compounded 15-20% growth every year and it doesn’t seem it will stop anytime soon whereas Laptops and Desktops are on a downward scale.
Once the basic understanding was given, he then moved onto RPI900 which is the immediate solution to have self-sustained networks in villages. What he did was to have two nodes, one attached to an RPI900 module with antennae (attached to a system) and the other a wi-fi antennae of a laptop attached to a Raspberry Pi as well and both were communicating over wireless. We were able to see networking established with both nodes being able to ping each other that and this was all in the freely licensed 2.4 GHZ .
Now the idea behind the demo was and is to get people excited. The RPi is an immediate solution but not the best solution as Raspberry Pi itself is partly closed-source, the 3d acceleration bit along with some other bits. The reason Raspberry Pi Model B was used is because it is the fastest yet, hence if Beaglebone Black (BBB) or some other open source hardware comes with similar power and at the same price range, probably hamara would move to that. There are quite a few black boxes within Pi which we are not happy with and we would want to change with having an open-source version as and when it comes feasible and available, we would like to move to that. The more open-source the hardware, the easier it is troubleshoot and design/dedicate manpower for mainstream as well as corner cases. Also having an open-source hardware will give us much more flexibility to do programming and maybe even make boxes for different purposes.
The idea of having a cheap open-source long-range wireless device is that at least some troubleshooting will and can happen locally without getting the devices to cities and no need for expensive support. Only when we could produce the whole thing locally or at least majorly can we expect to have Economies of scale. This is a short to medium-term project and we (as in Hamara) would welcome any contributions that people could make in making this dream a reality.
If this is something you would like to take up, head to either our IRC chat , take a nickname and go to #hamara or/and join the Hamara Developers Mailing List . Make sure to introduce yourself and whatever ideas you have.