Before starting any business in Gaza or any place in the world, you should study the available local resources. Gaza lacks a lot of things; electricity, water, food, transportation, etc…. But it very rich with something unique; its people. Yes, the type of Palestinian people is very unique. Why is it unique? because these people are born to be engineers, how come? well, everything is a challenge and any one wants to survive has to come with brilliant ideas that overcome all the barriers we have. For example, if you want to do simple thing like cooking go for a Palestinian home in Gaza and see home many alternatives they have! they cook on gas, electricity, wood, old cloths if available, gasoline, and many other things I cannot count. This is just to cook a nice dinner. Now these people not only came with more than one solution for the problem, which is an essential part of being an engineer, they also making trade-offs on a daily basis. Let’s go back to our cooking example, every Gazian home has at least three types of the cooking methods I mentioned and for each meal on the day the family has make a trade-off on which method to use to cook. The choice is based on three factors: Availability, Cost, and Time. These are the principles the people go for four or five years college to learn but we learnt by practice sine we are babies.

From all this and many other every Palestinian learn to be an engineer with a great passion, determination and no fear from failure. If someone failed in doing something, giving up is not an option but trying again and again with learning valuable lessons form each failure so we become better on every new day.

I think that what makes the Palestinian people one of the best educated people in the region, despite all what they are facing. For example now in Gaza people have electricity for less than half of the day in average, but the internet penetration rate is 60%!

I think the best product that we can produce in Gaza is innovation! yes innovation is a very valuable product and it’s more rare than gold! IT industry are the best match for marketing innovation. It doesn’t need huge resources nor capital of investment but it needs brilliant well trained people.

Now back to my friends question, he was really excited by Google’s visit to Gaza and he wanted to do something related to their products so I limited my answer to areas were Google products can be used (Maybe in the future I will blog about a broader business chances in Palestine). Here is and edited version of my answer to him

For the ideas, I cannot tell you specific ideas but I can help by giving some thoughts on directions where to invest:

1. Focus on small ideas that target community needs, you cannot change the world with on application. By this I mean think about things that people would use everyday, for example small Android apps that organize your daily consumption of calories are successful though it’s very simple (I’m not telling you to write app for daily consumption of calories but just an example)
2. Google now focusing on having more market share on the Internet advertisement on the Middle East, I heard that advertisements market in the middle is 10 billion dollars per year while internet advertisement is only 100 million dollars. So any idea related to advertisement on the Middle East they will be interested on.
3. Try to think about integrated experience, like apps that the user would use from a smart phone while going to his work or school, then when he is at his office he will use the same app but from a desktop, and when he is at home he will use the same app from his HD TV (look at Google TV).  The technology to build such applications is not that hard, but the hard part is coming with ideas and apps that the user want to use on the first place. Take social networking as an example, everyone interested to update their status and check out their friends status updates thats why people having Facebook app everywhere from their cellphones to their desktop and now even on their TVes.
4. Expensive product doesn’t work. To get money you have two options. First making your product for free and using advertisements as source for revenue. This is more suitable for web applications targeting any kind of devices (cellphones, TVes, Desktops). Second, is having more than one small app and selling it for low prices (less than $2) but focusing on selling a very large number of it.
5. Think global, Gaza is not a big market but we have access to the internet that’s mean you can sell your application to anyone on the world. This sounds scary sometimes for startups in Gaza but they can do it only if they belived in themselfs.

Hopefully this friend and others succeeded in their efforts in making Palestine a better place. Again, I’m not an expert I’m just another Palestinian guy who have an opinion about the subject.

1. Inherently sequential code can be improved by only small factor. Thus no need to put too much effort to make it 100% parallel, any small improvement in running sequential code is a good achievement.
2. If a good parallel programming model is provided, software developers will eventually start writing more parallel code. This again minimize the need to make the sequential code run in parallel.
3. Any future designs should address both fine-grain and coarse-grain parallelism. Addressing both types of parallelism will give the flexibility for the programmer to exploit the maximum parallelism on the application. But it would be better not to let the programmer worry about what type of parallelism, compilers, operating systems, and hardware should take care of it.
4. Von Neumann architecture  use only one memory port for both data and code, while in Harvard Architecture it uses a port for data and another port for code. While current modern architectures, even those who started as pure Von Neumman,  are using mixed of both architectures. For example Intel L1 cache is divided to both data and code cache, very similar to Harvard Architecture,  while there is only one main memory for both data and code, like Von Neumann. But having 1000s of cores on the same chip will require more memory ports. By more memory ports I don’t using Harvard architecture or building systems similar to NUMAs. We need to figure out a smarter memory schemes that fulfill the needs for 1000s taking in mind that the advancement rate on the memory speed is much less than CPUes.
5. With all this cores per chip and the amount of the available parallelism, concurrency is going to be a major concern and efficient standardized methods is needed to handle this issue. Handling concurrency should be embedded deep on the stack from the hardware to the operating system to the programming language so the end programmer shouldn’t worry that much about it. Of course it’s not going to be totally free for the programmer to use concurrency, but at least should be very easy, pain free, and efficient.
6. Another thing that I already talked about in my previous post. We need to rethink about the contract between the different layers of the stack that we built so far. Thinking again is required on each layer as well as on the  contract between the layers.

Most tutorials I saw on the Internet were talking about how to install a grub on a physical floppy. Of course no one these days use floppy disks not they have floppy drives on their machines on the first place. This tutorial is inspired by an OSDev article http://wiki.osdev.org/GRUB#Installing_to_floppy.

Creating a grub boot loader disk, two stages are needed. First stage is creating an auxiliary disk. The auxiliary disk is needed to copy stage1 and stage2 of the grub boot loader[2] in block 1 and block2 of the disk. This means we have a bootable disk but the filesystem of the floppy is corrupted. We need a floppy with a fully functional filesystem that we can use to copy our kernel on it. Second step is create the actual boot floppy with a fully functional filesystem. The second floppy is made by using the auxiliary disk. Enough talking and lets go to the practical steps.

Stage 1: Creating Auxiliary Virtual Floppy

1. The  very first thing is is to create the virtual image that we are going to work on.

dd bs=512 count=2880 if=/dev/zero of=auxiliary.img

Basically, this command will create a file of 2880 block each one is 512 bytes (remember floppy disk is 1.44MB). Because we are reading from /dev/zero the new file will be initialized to zeros.

2. Copy stage1 and stage2 to the image. This is done with the dd command. But things we have to be aware of. First, stage1 should be in the first block of the disk and stage2 in the second block. Second, dd by default with truncate the floppy image, something definitely we don’t want, so make sure to use conv=notrunc option.

dd if=stage1 of=auxiliary.img bs=512 count=1 conv=notrunc
dd if=stage2 of=auxiliary.img bs=512 seek=1 conv=notrunc

You can download a ready made disk from here: auxiliary.img.tar

Stage 2: Making the actual boot disk

1. Same as first stage, we need to create the image file initialized to zeros.

dd bs=512 count=2880 if=/dev/zero of=grubboot.img

2. (Not necessary step, but I like it), Attach the image file that we just creating to a loop device.

sudo losetup /dev/loop1 grubboot.img

By this way we will work our image as an actual block device, rather than just a file. I used this for two reasons. First it’s more convenient to handle block device. Second, when making a file system on a file using mkfs it will complaining about it not being a block device, but still works!

3. Make a filesystem on the boot disk and then mount it. You can choose what ever filesystem you want as long as it’s supported by GRUB!

sudo mkfs /dev/loop1
sudo mount /dev/loop1 /tmp/virtualfloppy

4. Once the filesystem is created and mounted we can play with this little floppy as we want. Now it’s time to create  a boot folder and add stage1 and stage2 to it. The other thing we need is a configuration file for grub menu. This file basically tells grub what operating systems exists and from where to load it. Of course to forget to copy your kernel as well. Here I chose to name by kernel kernel.bin and put it on the boot folder.

sudo mkdir /tmp/virtualfloppy/boot
sudo chmod o+w /tmp/virtualfloppy/boot
sudo cp stage1 /tmp/virtualfloppy/boot
sudo cp stage2 /tmp/virtualfloppy/boot
echo "title   MyOS
root    (fd0)
kernel  /boot/kernel.bin" > tmp/boot/menu.cfg

5. We are almost done. Right now we have a floppy with a filesystem and all our files, but not the boot sector. And here comes the role of the auxiliary disk we created on the first stage. First boot using the auxiliary by any virtual machine software you like (I use VirtualBox[3]). If everything is working good you will see a grub prompt on the virtual machine. Now unmount the auxiliary.img and mount the grubboot.img. After doing this type the following command on the grub prompt.

install (fd0)/boot/stage1 (fd0) (fd0)/boot/stage2 (fd0)/boot/menu.cfg

6. And we are done. Just reboot from grubboot.img and your kernel will be running. One thing worth to mention here, this is a really long process to create a virtual bootable floppy but it’s worth it, because once you created the floppy you just copy your kernel again and again without needing to repeat anything from the previous steps. The other thing to mention is, I’m using the legacy GRUB here because I don’t really need the new features on the newer versions of grub I just need something that boots my kernel.

And if you need just a floppy that works by copying your kernel into without going through all steps, download this ready made image but without a kernel: bootfloppy.img.tar

[1] http://www.gnu.org/software/grub/manual/multiboot/multiboot.html

[2] Probably you can find these files installed into your linux, but if not you can download them from here grubstages.tar.

[3] VirtualBox http://www.virtualbox.org/

The current modern systems are built in layers and layers of abstraction. The idea behind abstraction is to divide complex systems into layers and each layer has relatively simpler complexity. But we are accumulating on the same layers built one the beginning of the computer age. For example the microprocessor is the same as the first microprocessors built. Some people might argue that and I agree with them we have very advanced microprocessors today comparing to the Intel 8086. But here I’m not talking about the technology, I’m talking about abstraction. Computer designers are using the same interface for the CPU level of abstraction since the creation of the first microprocessors.

This huge accumulation of old levels of abstraction will start failing in keeping serving the current needs for today’s and future’s technology. One example of that is the multicore processors. With the introduction of multicore we start hearing from researchers that we need to rethink of processors, operating systems, and programming paradigms. However, most of the current efforts directed to keep the same contract between the different layers of abstraction and just change the internal of each layer without breaking the contract between the layers.

I think these levels of abstraction are not written in stone and we don’t just need to think about improving each layer, but we need as well thinking of the abstraction levels from time to time. However, I don’t recommend changing the abstraction levels very frequently. But I do think that there are some times where we need to rethink of the whole stack, and today is one of these times.

In rethinking of the abstraction levels there are two approaches. First approach is rebuilding the whole stack again from ground up. This approach gives the designer total freedom of building new stack. One particular project that I like and follows somehow similar approach is the Par Lab at Berkeley [1]. Researchers there are aiming to setting new agenda for multicore research starting from anticipating the future applications and then design new stack the will support these applications.

The other approach of the rethinking of the stack is to use ideas from higher levels of the stack in the bottom of the stack, or vice versa. This approach allows the designer to reuse already well defined and tested solutions. One project I liked that follows the same approach is Factored Operating System (fos) at MIT [2]. They are using the same concepts Internet to design operating system that is service based. I liked this idea very much since I was thinking of it for more than two years. At the beginning I thought that is very crazy idea but not it’s real.

[1] http://parlab.eecs.berkeley.edu/

[2] http://groups.csail.mit.edu/carbon/?page_id=39

Good luck for everyone….

[0] http://groups.google.com/group/google-summer-of-code-discuss/browse_thread/thread/d839c0b02ac15b3f

[1] http://spreadsheets.google.com/pub?key=p6DuoA2lJToKmUzoSq6raZQ&gid=1

http://www.udel.edu/udaily/2010/dec/supercomputing120409.html ]]> http://www.hassany.ps/2009/12/sc09-competition/feed/ 0 http://www.hassany.ps/2009/10/working-for-mog-more-about-the-project/ http://www.hassany.ps/2009/10/working-for-mog-more-about-the-project/#comments Thu, 08 Oct 2009 04:22:37 +0000 ahassany http://www.hassany.ps/?p=71 If you from more advanced country it’s simple to send mail because there is a system for addressing, but this is not the case about Gaza and this one of the reasons that the mail never works in Gaza. Therefore, Municipality of Gaza (MoG) started a project funded by the world bank to make system for addressing. The idea is very simple, just give a number to each street and a number for each building on that street. By this way each building will have unique address by the pair (BuildingNo, StreetNo). But implementing this on the ground it not easy as paper. It consumes large human resources because Gaza is old city and it not very well designed city and in many cases people working on that project spend days just thinking how normalize special cases that arise. But this only one part of the project.

My project was the second part of the project. Since every building must have at least one construction permit this mean that each (BuildingNo, StreetNo) pair must point to one or more construction permits. One or more because the building can be built in different phases and each phase must have separate construction permit.

MoG have big old archive of construction permits. This archive contain permits since 1910s but this archive is paper based and no computer records about it. We are now in 2009 and MoG still have no computer system for construction permits all is done manually.

My boss give me a three simple tasks:

1- Get the important data from the paper archive to computer database.

2-Match each construction permit to (BuildingNo, StreetNo).

3-Build new computer based system for future construction permit where each construction permit will automatically attached to (BuildingNo, StreetNo).

These tasks seems very concise and simple!

Next time I will talk more about the nature of the problem and the approaches I took to solve it.

During this periods I work on big project and I was responsible of big part of it. I gained a lot of experience and mostly non-technical experience. Things like how to manage team and how to collect business requirements from people know nothing about computers.

I will try on the next few weeks to summarize what I learned on this blog so I don’t forget what I learn and share this experience with others (of course, no one has to work the same amount I worked )

I hope that I will do well in my new life.

Are eager to learn and work with amazing people from around the world?

Do love to work in the benefit to humankind?

Do you want to know how is participating in open source development will be in benefit of you own professional career?

To get answers for all these questions and more, join me in my workshop during the Engineering and Technology Meeting at IUG on May 20^th,2009 11:00AM – 12:30PM.

For more inofrmations about  Engineering and Technology Meeting at IUG visit: http://prl.iugaza.edu.ps/ex/

To make time line I usually do the following:
1- Define from where I will start  and what the end result I want to achieve.
2- See what tools and technologies I will have to learn in order to achieve my goal.
3- Define the project architecture. It’s good practice that my project architecture follow a well known design pattern like Model-View-Controller (MVC) or whatever is suitable for my project. But if I can’t make my project follow a well known pattern it’s OK design pattterns are not for all type of project. But always try to make your architecture into layers and modules.
4- Divide my big goals into millstones. Each millstone  means I have something that can work and can be tested. For example, if I’m following Model-View-Controller (MVC) design Pattern then finishing the model is a millstone and finishing the view is a millstone, also finishing the controller is a millstone, testing is millstone, packaging is a millstone, etc..
5- On each millstone I try to divide it to a smaller tasks like I want to implement class A on day1 and class B on day2, I want to test module X on day 3, etc.
6- Always put some time for catchup in case I didn’t follow the timetable. GSoC is very short programm so I guss you can’t put more than week for catch up.
7- Always Revise the time line you put on the proposal during the Community Bonding period because once I know more about my project I will be able define more accurate time line.
8- Always do weekly progress reports even I don’t made any progress. By this way I always put the time table on mind and do my dest to follow it.

1. Go through the mentoring organizations list on http://code.google.com/soc and compare it to you preliminary list.
2. You may find some organizations in your list but it not accepted by Google, Exclude it from your list.
3. Or you may find other organizations are listed you’re more interested on. You can add it to your list. But remember this is semi-final list so don’t make it big I don’t recommend list bigger than five projects and try to select just one project from each organization.
4. It’s OK if all the preliminary list is changed. The goal of the preliminary list is to get you on the mode. But I don’t recommend excluding projects that you send patches to it and these patches where accepted because this mean you already get high possibility to be accepted on that project.
5. Ask the contacts for each project for more information about the project.But don’t ask trivial questions!. Personally I prefer not to ask “What I should go?” or “Can you please give more information about project X?” because this questions show that I don’t understand anything from the listed Idea and I’m waiting the potential mentor to write the proposal for me. But I prefer to think on the idea and put a vision for the project I want to ask about it then go and ask the potential mentor “I have though of doing 1,2,3 for project X” and then ask really smart question about things that was unclear to me.
6. Some project might seem unclear or ambiguous. This might be for two reasons. First, some of the project are intended to be ambiguous because the mentoring organization wants to reveal your innovation. Second, you might don’t have previous experience on the project subject; it’s OK we all here are students and we came to learn new staff, so try to read more about the project and give it some time to think about it. If you have a vision for the project send it to the potential mentor and ask him for his comments, otherwise if you couldn’t put a vision for the project look for another one and don’t worry their is a dozen of projects! But always remember time is critical don’t stuck on one project there is hundreds of ideas and some of it maybe better than you’re currently stuck on.
7. For the ideas on you list try to start put a vision for the projects how it going to be implemented during the summer.
8. Start working on the proposals. It really important if the organization got a template for the proposal that you follow it, but if it don’t have a template try to make your proposal similar to a good template from other organization.
9. Also as the previous stage poke around the mailing list or hang on the IRC channel of the mentoring organizations you are interested on.
10. If the organizations on you short list has created mailing list or IRC channel specially for GSoC  do subscribe on it and use it to ask your *smart question* and to get know the potential mentors.

Good luck everyone with GSoC. It’s competitive and a lot of work but it’s really fun and you will remember these days for the rest of your life.

You can find the list of the accepted metoring organizations on http://socghop.appspot.com/program/accepted_orgs/google/gsoc2009?limit_0=100&limit_1=100

MySQL GSoC 2008 Logo

Free Ubuntu CD

The seconad was sunny! and I did better than the first day. On the second day I talk specificlly on GSoC and how to join this program. But I notice that some of the students have some fears from participating such big program some of them might think “Oh, this is Google and the most prestigious FOSS organizations what I’m going to do! “. I belive GSoC is a challenging program in which I pushed my self to the limit. GSoC is designed for students and they know that you’re a student and you’re here to have fun and learn.

GSoC Statistics

I’m proud to join GSoC 2008. GSoC was a great experience for me and it really changed my life. I’m writing this post to give the new GSoCer some tips that I hope it will be useful for them. In this series I will give different tips for all the stages of Google Summer of Code™.

Google Summer of Code™ 2009 Logo

Stage One: Before Google Announce the Mentoring Organizations (from now to March 13)

1. Look at the previous GSoC mentoring organizations, most of them has already published this year ideas list. You might try using Google search engine to search for ‘Google Summer of Code 2009 Ideas’.
2. Try thinking of those ideas, see what might interest you. I would remind you that GSoC is about three to four months you don’t have much time for big projects. Also, being ambitious is good but remember you have to be realistic about your project don’t over take projects that are over your capabilities. In GSoC you’re going to learn a lot of things and you mostly you will not have all the knowledge and experience to do the project. But you should got the basics and you have all the summer to learn and work. So, the best thing is to try to balance between what you already have and between what you’re going to learn.
3. Make short list about 5 to 7 projects from different organizations. You free to select more, but big list will distract you. Also, try to select projects that have similarities on your advantage, like all the project on your list use PHP and you have PHP experience.
4. Start walking around on the mailing lists and IRC channels for these projects.
5. It will rise your chance to be accepted if you send patches for those projects, but if you don’t it’s OK.
6. If you feel that those projects on the short list is hard and over your capabilities don’t panic. Try thinking again on that project sometimes project look very hard at the first time but when you think of it or discuss it with your friends you will see how easy it is. If you think of the project one time after another and you still find it very hard it’s OK, try looking for other projects.
7. Note: not all organizations that apply for GSoC will be accepted. It’s also competitive for the mentoring organizations, so be sure that your short list contain projects from at least three diffrent mentoring organizations.

The First day I will introduce the Open Source and the philosophy behind it. On the second day, I will talk about Google Summer of Code. And on the third day I will talk about Open Source developments and the things that will be useful for students to learn before  joining Google Summer of Code.

Google Summer of Code™ Your Way to a Global Mission Workshop Invitation

Note: Each attandant student will get a free copy of Ubuntu Linux from PNINA and will get GSoC sticker from Google.