Archive for July, 2009
NZ – Second day in University – Contemporary issues
Posted by JohannesTheLittleScientist in NZ, serious business on July 21st, 2009
I’m just blogging these posts, because a lot of people don’t know what computer scientists do. “It has to do something with computers, they sit in front of it and write numbers all day.” I would find that quite boring.
I think I came up with a good definition of what Software Engineering is about, just before I got my bachelor in Software & Information Engineering. It is about getting into a “domain”, for example the car industry, understanding their terms, processes, work, information, problems. Then coming up with a model that describes or solves an issue. The implementation is usually left to programmers. I left out some details about maintenance, that it is an iterative process, etc.
So “Contemporary issues”. Every master degree has that I think. It is about finding out what people are currently researching in. Now computer science is a broad field, I’ll give you two examples, areas I’m interested in doing the paper about:
a) Representation of mathematical theorems and automatic proofs
Probably very theoretical. Some websites in that area are vdash.org and metamath.org. It is about “teaching” the computer our understanding of math. This works very well for logic, algebra and geometry. But what about analysis and calculus?
b) Swarm algorithms, (maybe with a focus on inhomogeneous and not connected all the time)
Swarm algorithms is about hundreds of small devices that can move and talk to each other. There is (normally) no leader, but a swarm algorithm (which runs on each device) emerges a behaviour of the group. My favorite example is a shark hunting through a swarm of fish. Each fish is just one fish, trying to avoid the shark. But you can also see the swarm showing some behaviour as a collective. Devices might be flying and exploring an area using sensors and telling the others which places to avoid.
The computer is just one side effect of computer science. There are theoretical areas that might not go beyond paper and pencil. Then of course there are all the combinations with other fields (bio, geo, chem), graphics, games, and effects on society as well as business related stuff.
NZ – First day in University: Ubiquitous computing
Posted by JohannesTheHuman in NZ, cool project ideas on July 20th, 2009
First day in University started just as the University in Vienna: Sleeping until 12, breakfast, then show up at some point for the lecture (3pm). Well, they’ll move it to 9am, so that’s that.
Ubiquitous computing: Small computers you use but aren’t necessary aware of. They are just there, all the time, carryable (or not), multipurpose. PDAs and mobile phones are the more obvious ones.
For a research project theme I’m thinking about doing research for store-and-forward protocols using inhomogenous, non-persistent networks.
Imagine walking through a busy street, people having mobile phones, PDAs, etc. Imagine you would have fast means of connecting these devices (e.g. BlueTooth, Wireless) and send short messages in the seconds while people pass by you. A message can hop from device to device to ultimately reach its destination (a phone, the internet, …).
Think of people making the net rather then the net coming from outside and people being attached to it.
For the technical side: Surely only short messages are of interest. But these can be any kind of data (e.g. TXT aka SMS, pictures). Every phone can yell “I am 123″, “I am looking for 234″. Assuming phones have some storage, the devices can build up a tree of phones they have seen (some very dynamic routing protocol). Messages can then be passed (or copied) in the hope they’ll reach the destination in this direction.
For the security side: The data may be encrypted. The phone numbers for routing can be hashed.
But think of a twitter network available even after an earthquake destroyed the mobile network transmitters. Some devices may support GPS, so positions can be approximately recorded. This might ease finding people (if they want to be found). If the net gets sufficiently dense (think Manhattan), a really efficient network might emerge where you can reach people in real-time.
Obviously these thoughts have issues, but I’m interested in what research was done by people in this area. I’ll not make research on this myself, and probably not doing field-testing or programming (well maybe routing protocols on simulators).
PdfJoin for Windows
Posted by JohannesTheDeveloper in Happy Hacking on July 19th, 2009
Allows you to merge and split PDF files.
Download installer pdfjoin-setup.exe
Try it out and recommend it to others.
PS: This is just a Windows distribution of an earlier post, packaging GTK+ and Ghostscript using py2exe and innosetup. GTK+ is licensed under GNU LGPL 2.1, Ghostscript is licensed under GPL 2, and the GUI under a modified BSD license.
Recent Comments