Very exciting news. I received bridge funding for my PACC research grant, and we will now start the assoicated interships with the respective students. I already welcome Alexandr, Alxandra, Temirlan, and Assulan, who will now enrich this blog with their wisdom.

We have discussed the following noteworthy points today in our first meeting:

• Every student blogs (you receive a blog account on ulno.net in the first meeting) on Monday first thing they do (do not spent more than an hour, be very coarse and informal, blog both successes AND failures to encourage comments from your peers)

• After this, read blogs of peers, feedback, (on some time on Monday, praise, encouragements, ideas)

• Wednesday 13:00 will be weekly meetings, everybody raises successes and failures again, we will discuss next steps and collaboration (everybody can call a meeting when stuck or necessary to allow integration)

• Every student shares a timesheet on google docs with Assulan and me, this is your initial commitment document, with your work forecast. You will fill in your real work hours every day with a small note about content (this helps you also blogging on Mondays)

• The lab is open for you all the time, make sure to talk to each other to share the limited amoutns of keys or make sure that either Assulan or me are in the university to let you in or out

• You should be at least around 70% of your time present in the lab but you are welcome to also do some research work at home (just note this as home office in your work sheet)

Today is the starting day of the internship of Cloud Computing project. I will be working for about 6 hours per day for 6-8 weeks.

I am going to start with working on iTALC.

I am also interested in working on setting up remote desktop solutions and finding out the most appropriate one (VNC, NX, etc).

Hi,

My name is Alexandr, and I'm an intern at NU for Cloud Computing programm. My workhours is from 09:00 - 13:45

Topics of research:

1.

Remote file/app pushing/Deployment

2.

WOL

3.

Cybersecurity

4.

Transparent proxy

5.

italk

Stay connected...

Hi,

My name is Assulan and I am a Teaching Assistant in the Department of Computer Science in Nazarbayev University. I will be involved in PACC project.

• Cloud Computing Contracts

The new delivery model for information processing, Cloud Computing is offering an attractive value proposition: Dynamic (on-demand) provisioning of IT resources Low cost because of the use of economy of scale and multi-tenancy sharing of resources Shorter time to market in deploying resources Different cost structure; the shift from CAPEX to OPEX But although this [...]

• Disable all swapping on Ubuntu Linux

That's how it goes for Ubuntu:

Disable current swapping strategy ( swappiness it's called)

echo 0 > /proc/sys/vm/swappiness

Disable swapping all together

swapoff -a

Don't swap after restart as well

echo "vm/swappiness=0" >> /etc/sysctl.conf

P.S! Read man-pages first and use at your own risk!

• My Search Logger is OUT!

Hi all!

I am happy to announce that my proxy-based search task logging software is now (finally) available to DOWNLOAD! There are a number of new features and improved existing ones. In order to get an idea of what to expect, I will share a few snapshots. Happy searching!

• 5 Ways Your Users Can Benefit from Cloud Anti-Spam

While many email admins prefer to keep their messaging hygiene in their own datacenters, others who are looking to save money and reduce the amount of hardware in their datacenters should consider cloud anti-spam solutions. There are some great features in cloud anti-spam that can help admins and their systems, and, let’s not forget, our [...]

• Rolling your own watchdog for Qemu

Sometimes there may arise a need to make your own watchdog for your virtual machine. The process itself is really a simple one, I will base this example on Intel's 6300esb watchdog.

Let's first start by getting the latest source code for Qemu:

git clone git://git.qemu.org/qemu.git && cd qemu

Now it is reasonable to switch to your own git branch in case you want to roll out some patches later on.

git checkout -b custom-watchdog

All of the existing watchdog sources are located at hw folder.

Make a copy of existing watchdog and include it in the build path.

cp hw/wdt_i6300esb.c hw/wdt_custom_watchdog.c

Now lets plug the new watchdog into the build process. For doing that we need to edit Makefile.objs and add a new line hw-obj-$(CONFIG_PCI) += wdt_custom_watchdog.o

right after the "hw-obj-$(CONFIG_PCI) += wdt_i6300esb.o" line.

Next step is to dig into the code and start modifying the new and improved watchdog. For the sake of simplicity I will leave the core functionality unchanged. I will only change the important parts of the code to enable the registering the new watchdog with Qemu. Replace every reference to i3600esb and alike with your version of the name. If you skip some of the code then there is a chance that custom watchdog will not behave as expected. Most of the tedious replacement is easily done with sed commands similar to this.

sed -e 's/i6300esb/custom_watchdog/g' hw/wdt_i6300esb.c > hw/wdt_custom_watchdog.c

Now open up the code of the custom watchdog and fill in it's name and description with your own such as:

.wdt_description = "Super Custom Watchdog"

Recompile now and see if the new watchdog is visible and loading with Qemu.

As it turns out, the new watchdog is visible for Qemu.

Let's run some VM with our new watchdog:

./i386-softmmu/qemu-system-i386 -watchdog custom_watchdog ~/debian.qcow2 

If everything goes well and the virtual machine comes up without any errors regarding the watchdog, the next step would be to replace some of the old variable names in the source code with more appropriate ones, otherwise the new watchdog is ready for further customization.

Source code here.

Software versions used:

• Qemu sources for  version 1.0

• How to do a search report? The Search Logger 2.0 usability study

*Summary*

*The background*

The aim of the Search Task Repository project is to let the users/Internet searchers publish their annotated search logs and make these logs searchable. In order to log ones search activity and behaviour, we have developed a special tool called The Search Logger 2.0. Here "2.0" means that a conceptually similar but technically very different logger existed before I started with this project and was developed by Hannu Kikkas. The current version is platform-independent and uses a proxy to modify every web page served so that it could be logged by our dedicated HTTP server. The Search Task Repository is currently in development and will be one of the main topics of my Master thesis.

** **

The goal

We wanted to put ourselves in the end user's shoes and evaluate the usability of the current Search Logger 2.0. To do that I conducted a number of exploratory search tasks and used our software to log the web activity. In essence, I sought an answer to the question "How to do a search report?". The other goal was to find, document and fix bugs in the existing logging and reporting software and come up with ideas about new features.

** **

The results

The existing solution was tested on six exploratory search tasks such as finding a foreign university on a budget where to study machine engineering, finding 5 clever Christmas present ideas for under 10 euros per item, finding a ballroom dance school in Tartu area, etc. These tasks and how they worked out on our current software in more detail can be found on my blog. What is more, I made a short screencast for presentation purposes which is also linked to. During the semester I made some small tweaks to the Search Logger 2.0 but they are mostly bug-fixes and I do not consider them significant in this context. Here is a list of things that I learnt from the experiment:

• FIX: Amazon book preview does not work properly
• FIX: Sidepane video image previews on YouTube are not displayed
• FIX: Some random exceptions (documented) -> possibly related to the pre- and post-forms

* TODO: Users might want to take notes while searching
* TODO: onSubmit and onClick events should be also logged
* TODO: The logger is lacking a database integration
* TODO: Search logger and repository integration (work in progress)
* IDEA: Copy-paste detection and bookmarking would be nice to have because this help us detect web pages with relevant information




Thank you for reading this blog and feel free to share your ideas!




Peeter




![](https://blogger.googleusercontent.com/tracker/6354410659114914157-3041369450356581481?l=searchtaskrepo.blogspot.com)





* [Finding a ballroom dance school in Tartu area](http://searchtaskrepo.blogspot.com/2011/12/finding-ballroom-dance-school-in-tartu.html)


And here is the second search task conducted. What makes it special is its locality. Search tasks often get a different meaning when the physical location comes into play. And this is one key aspect to bear in mind for developing the repository of search tasks. A user who is searching for a dance school in Tartu might have little use of search logs describing the same thing for Stockholm. However, general search patterns (for instance some non-location key words typed in Google which lead to good results) might still be discovered from the logs.

Pre-form:


[![](http://3.bp.blogspot.com/-p38O1t5PBMU/TuhnqAJ0V0I/AAAAAAAAACE/Nle_YTlpv4w/s320/dance_pre.png)](http://3.bp.blogspot.com/-p38O1t5PBMU/TuhnqAJ0V0I/AAAAAAAAACE/Nle_YTlpv4w/s1600/dance_pre.png)


Post-form:


[![](http://4.bp.blogspot.com/-JcZd2uqyFmo/TuhntL7_JEI/AAAAAAAAACM/8AoVHPsBhZE/s320/dance_post.png)](http://4.bp.blogspot.com/-JcZd2uqyFmo/TuhntL7_JEI/AAAAAAAAACM/8AoVHPsBhZE/s1600/dance_post.png)


Technical logs:
[http://www.mediafire.com/?86v2dlqhp0qlvl9](http://www.mediafire.com/?86v2dlqhp0qlvl9)

In the next post I will wrap up why the things I did were useful for the project in general.

Peeter

P.S. This is not an advertisement but the best school quality/price-wise that I found was Revalia Tantsukool, appr. 4-5 EUR per hour. :D

![](https://blogger.googleusercontent.com/tracker/6354410659114914157-8206623779854483382?l=searchtaskrepo.blogspot.com)





* [Electronic books search case study](http://searchtaskrepo.blogspot.com/2011/12/electronic-books-search-case-study.html)


Hi!

In the meantime I have came up with two new exploratory search tasks. The first one will be covered in this post

Objective:


[![](http://4.bp.blogspot.com/-4_97YKwlWxI/TuhkhofLa8I/AAAAAAAAAB0/pSoB8h4pBOc/s320/books_pre.png)](http://4.bp.blogspot.com/-4_97YKwlWxI/TuhkhofLa8I/AAAAAAAAAB0/pSoB8h4pBOc/s1600/books_pre.png)


Result:


[![](http://4.bp.blogspot.com/-jqLR_0FoKkY/TuhklCIRSHI/AAAAAAAAAB8/G_BXLKplZRo/s320/books_post.png)](http://4.bp.blogspot.com/-jqLR_0FoKkY/TuhklCIRSHI/AAAAAAAAAB8/G_BXLKplZRo/s1600/books_post.png)

Answers:

Teach Yourself Electricity and Electronics [Paperback]

Make: Electronics (Learning by Discovery) [Paperback]

Tab Electronics Guide to Understanding Electricity and Electronics [Paperback]

Electronics Demystified [Paperback]

Electronics For Dummies [Paperback]

All New Electronics Self-Teaching Guide (Wiley Self Teaching Guides) [Paperback]




Search logger logs and crash logs:

[http://www.mediafire.com/?yzcyt02hqn9u5n6](http://www.mediafire.com/?yzcyt02hqn9u5n6)




There is more to come...




Peeter





![](https://blogger.googleusercontent.com/tracker/6354410659114914157-3686203593636295608?l=searchtaskrepo.blogspot.com)