-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
25 lines (21 loc) · 5.68 KB
/
index.html
File metadata and controls
25 lines (21 loc) · 5.68 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
<!DOCTYPE html>
<html>
<body>
<h1>My Github Blog</h1>
<br></br>
<div>Assignment 1 CMPT395</div>
<p>
Recently I found a very interesting developer of free and open source software for CMPT395. The developer I found was a Dr Linas Vepstas, who is the primary maintainer and developer of the OpenCog's linguist module(RelEx) and also assists OpenCog in many other respects. OpenCog is a non-profit organization dedicated to building a artificial general intelligence, the RelEx module is built to analyzes language structure in English. RelEx is used by OpenCog in the converting of English to machine code thereby enabling machines to understand what people are saying in essence. This is incredibly important in the development of a AGI, an artificial intelligence that can do multiple functions and can understand natural sentence structure of humans. RelEx allows for the understanding of context in English, when used in conjunction with the other parts of OpenCog. Linas has also worked for IBM, and has recently been working on the Linxus kernel for Power architecture-based mainframes which I discovered when reading his web-page http://linas.org/, he also has a blog about consciousness and machines on http://blog.opencog.org/2014/01/17/what-is-consciousness/ where he discusses if the implementation of consciousness in a machine is possible within the OpenCog's framework, he believes it is. I however think he is being too optimistic in the regards of developing an actual conscious machine, for the machine to be conscious as he outlines it not only has to hold a internal model of the external world, but also have an internal model of self, along with an internal model of attentional focus. These things would be incredibly difficult to implement all together, especially when you begin to provide sensory, motor and language input all at once. This is also why I believe that the OpenCog's time line on their website is highly unlikely to be met, due to the fact that they are predicting the development of an AI of human intelligence by the year 2019-2021. While I believe that the development of such an AI is possible, I think that it is highly unlikely to be developed within the next six years. I think it is far more likely that any AI's developed within that time frame, will excel at certain tasks such as either communication, or moving around in a dynamic environment but will not be able to do both. Human's intelligence is not just our ability to process numbers, or carry on a conversation, but it is a combinations of all the things we are capable of which makes it highly unlikely that a machine that is equal to mankind's' intellect will be developed. That said I'm still looking forward to the developments in the field of AI and who knows maybe I'm wrong, maybe they will develop a machine that is our equal within that time, if so i greatly anticipate seeing it in action.</p>
<p>Assignment 2.
For this assignment I attempted to work on a few different open source projects before getting one to function properly. I tried a few different open sources projects, the one I spent the most time on before settling on my final project(chess), was LibreOffice. I started b following the directions on their website for Ubuntu Machines:
Cloning it was easy the rest caused issues: $ git clone git://anongit.freedesktop.org/libreoffice/core. Building the dependencies caused particular problems: sudo apt-get build-dep libreoffice while this would run it did not on its own install all the necessary dependencies on my VM, at first this was due to a lack of space on my VM, which lead me to resize my VM's hard drive, which then broke the gnome windows manager I had installed previously for another open source project. This caused me to rebuild my VM from scratch and when I then tried to install the proper version of gnome that libreoffice requires sudo apt-get install libgnome-vfsmm-2.6-dev it failed due to the fact that deb-src http://archive.ubuntu.com/ubuntu/ precise main was not included within the /etc/apt/sources.list file for Ubuntu. After I then completed installing gnome and the other dependencies for LibreOffice I then got an error stating that a certain function was missing, after being unable to build it this was when i switched to another project. I later learned that the error statement was occurring because my VM was not a 64 bit version of Ubuntu.
Installing the Chess project was much easier, mostly because all of the relevant information for the project was on a single page that was consistently available, unlike LibreOffice which had its build instructions spread over multiple pages, which were constantly going down over reading week. Setting up chess was easy I simply forked the master branch and cloned it, set up upstream, and installed eclipse. The only problem I came across was after I had cloned the master branch I noticed that all new development was being done on the rewrite branch, so I now need to checkout that branch in order to work on the most recent version.
I also started confirming errors that i plan to work on below is a screen shot of one such error.</P>
<img src="https://github.com/WatkinsEric/WatkinsEric.github.io/blob/master/Chess_bug.png?raw=true"></img>
<p>This demonstrates that the function for calculating legal moves is currently not functioning properly.</p>
<p>Getting the rewrite branch to work is much more difficult then the master branch.
I had to install several different versions of java, in addition to installing IntelliJ IDEA, then had to config
the which java version IntelliJ IDEA was using for the project and now that i have the rewrite branch running it looks like
none they have solved numerous errors in their project without removing them from the issues page</p>
</body>
</html>