Monday, December 8, 2008

sya-710 lab 8 creating a live cd

livecd-tools is already installed on my f9 computer

I chose to use the minimal kickstart file to save time as it is
ok to update the system later.
I changed the line selinux --enforcing to --permissive in the ks file.

~]# livecd-creator -c /usr/share/livecd-tools/livecd-fedora-minimal.ks --fslabel=live-
fedora.min

Should end up with a file called live-fedora-min.iso
I'll return to this after the iso is created.

Friday, November 28, 2008

running wubi.py

still some issues - to start with i should have made a bigger windows partition when i installed xp

i slimmed down the isolist.ini file to one entry but i really need it to find the local .iso for figuring out how things are working - slowly getting further along though

below is the basic tasklist of what the installation consists of:
note that the tasks on the left match the message in the installation window
- helps just to get a basic idea of where things are


def get_installation_tasklist(self):
tasks = [
("Selecting target directory", self.select_target_dir),
("Creating the installation directories", self.create_dir_structure),
("Creating the uninstaller", self.create_uninstaller),
("Copying files", self.copy_installation_files),
("Retrieving the Metalin", self.get_metalink),
("Retrieving the ISO", self.get_iso),
("Extracting the kernel", self.extract_kernel),
("Uncompressing files", self.uncompress_files),
("Choosing disk sizes", self.choose_disk_sizes),
("Creating a preseed file", self.create_preseed),
("Adding a new bootlader entry", self.modify_bootloader),
("Creating the virtual disks", self.create_virtual_disks),
("Uncompressing files", self.uncompress_files),
("Ejecting the CD", self.eject_cd),
("Rebooting", self.reboot),
]






So backing up - in the basic info used before the windows frontend (the actual install window) the last function in the list below might be the one that finds the local .iso file

def fetch_basic_info(self):
'''
Basic information required by the application dispatcher select_task()
'''
self.info.original_exe = self.get_original_exe()
self.info.platform = self.get_platform()
self.info.osname = self.get_osname
self.info.language, self.info.encoding = self.get_language_encoding()
self.info.environment_variables = os.environ
self.info.arch = self.get_arch()
self.info.languages = self.get_languages()
self.info.distros = self.get_distros()
self.fetch_basic_os_specific_info()
self.info.uninstaller_path = self.get_uninstaller_path()
self.info.previous_targetdir = self.get_previous_targetdir()
self.info.previous_backupdir = self.get_previous_backupdir()
self.info.keyboard_layout = self.get_keyboard_layout()
self.info.total_memory_mb = self.get_total_memory_mb()
self.info.locale = self.get_locale(self.info.language)
self.info.cd_path, self.info.cd_distro = self.find_any_cd()
self.info.iso_path, self.info.iso_distro = None, None #self.find_any_iso()

understanding python's ConfigParser - sort-of

call these notes to myself: ConfigParser is a python core module that is used to parse a configuration file

ok that was a path issue - my test.py file was higher in the directory structure than wubi.py so it was able to find the data directory (i actually did two things - should have done them one at a time 1. modified pythonpath env variable to exhaustively include every directory being used 2. copied the data directory to the same location as the wubi.py file - i'll step back and see which one worked)

running wubi python code - the code uses a configuration file to both find out what to install and where to get it

the configuration file looks something like this - but with several sections (this shows only one section:
############ config.ini #####################
[mysection1]
arch=i386
name=fedora
packages=fedora-desktop
metalink=http://somewebsite.com/metalink
metalink2=http://someotherwebsite.com/metalink
ordering=1
##############################################

#!/usr/bin/env python
# the file to parse here is the actual .ini file shipped with wubi
import ConfigParser

myfiletoparse = 'c:\python25\scripts\intrepid.python\data\isolist.ini'
print('Parsing myfiletoparse=%s' % myfiletoparse)
configtree = ConfigParser.ConfigParser()
configtree.read(myfiletoparse) #reads file looking for [sections] etc

for distro in configtree.sections():
print('the section=%s' % distro)

##################### no problem parsing the file ###############33
the section=Xubuntu-amd64
the section=Kubuntu-amd64
the section=Ubuntu-amd64
the section=Ubuntu-i386
the section=Kubuntu-i386

current problem is that the file fails to open in the wubi code - exactly the same file, same path and everything. I added a print statement in the ConfigParser.read() function

#### part of Config.Parser.read() ################
read_ok = []
for filename in filenames:
try:
fp = open(filename)
except IOError:
print('cannot seem to open the file=%s' % filename)
continue
self._read(fp, filename)
fp.close()
read_ok.append(filename)
return read_ok
#######################################################################
here is the debugger output - not sure what is going on here unless it is apath issue
so i'll try the same fle in a different location or something

Wednesday, November 26, 2008

looking at the linux boot process


This is just a temporary record that I will pull together with some other stuff for figuring out wubi

Working primarily on Windows:
Extract files from the iso C:\7z.exe x F10-live-cd.iso

Then use Fedora to open up the ram disk:
lu@localhost$ gunzip < initrd0.img | cpio -i --make-directories
I copied it back to the windows computer to read the init boot script(s) and maybe edit them - then the whole lot (the stuff extracted from initrd.img) can be repacked
on Fedora using something like
lu@localhost$ find ./ -H newc -o > initrd.cpio
lu@localhost$ gzip initrd.cpio
lu@localhost$ mv initrd.cpio.gz initrd0.img

Note: this is a bit unweildy; maybe there is an easier way - but I cannot find anything that runs in Windows to take extract the files from a .img file but seeing it is really a gzipped cpio archive that might be possible (to do: rename the .img file to .cpio.gz and see if common windows extractors will understand it)

GNUwin32 does have gzip.exe, gunzip.exe and cpio.exe and they have no problem extracting the directory structure from initrd.img (so long as i name in .cpio.gz so that makes things a bit less unweildy - haven't tried torepack them yet but that is a detail for later - back to stepping through the wubi python stuff.

(That image - probably should say picture here - might be too small to read but...) the directory that is open in the image is the extracted initrd.img file and in the subdirectory sbin where the real-init boot script lives (after making any changes the whole iso will need to be zipped up again)

Friday, October 31, 2008

SPR720 reviewing

Reviewing a Fedora package.

NAD710 lab 5

A bit about SAMBA and xinetd ...

Once again Professor Chan took us through a 'hands-on' installation and configuration of an industry standard software product within the very brief two hour lab period for our networking course. He does this every time and while responding to both related and unrelated questions and coaching individual students in anything network related. His ability to deal with distractions while maintaining the direction of his students' focus toward the desired goals is remarkable - almost as remarkable as his mastery of the subject he teaches.

As sys admins samba is something we need to be aware of. At one time it would have been mandatory to have a working kowledge of this hybrid network product (hybrid in the sense that it is a link between Linux and Windows). Even though modern windows systems have moved away from the model where this product works so well, it remains a viable simple solution for organizing older networks that remain in place in some sectors due to lack of funding for newer equipment. I am not going to say anything on the details of samba just a link to the samba site.samba

xinetd is an interesting daemon. It is an interesting approach to managing server process in an uniform way. Choosing to introduce this 'server of servers' in the samba lab really drives home the point of what it is for. The real beauty of this lab lies in the fact that I came away having completed installations and configurations such that if I never look back at them, I still have a solid general and practical understanding/experience to draw on in amy real work situation.

In short, I have had a chance to reflect on the learning experience in my program and find the Lecture/Lab combination a really powerful combination in that we are given enough knowledge and practical experience to easily embrace a subject and take it as far as our own needs dictate. It leaves me with a sense of accomplishment.

SYA710 automated installation and breaking barriers

I'll fill this in later but a few headings now

1. isomaster
2. kickstart
3. trying to figure out boot loaders

isomaster - I thought of an .ISO file as a finished product something that could be copied but not altered. That is my misunderstanding. This lab had the intention of having us automate a linux installation. One side-effect was an introduction to isomaster and using it to modify our iso file to automate the installation. This tool is very powerful.

kickstart - this is the Fedora way to set up an automated installation. As I learn about Wubi (Ubuntu installer for windows), and its 'preseeding' approach to automating a linux installation, I see the kickstart approach as parallel thinking in different open software communities.

The lab had us download an iso file for fedora installation and modify the generic kickstart file that is created specific to each installation to copy it to the iso: thus creating a new iso that repeats our basic install with need for human intervention. This is one of those barrier breaking moments: a very enabling lab.

boot loaders - sitll a mystery to me!!

wubi weekend

1. figure out how to build wubi - i.e. come to grips with the python installer and its makefile setup
2. build it without making it compiled code
3. single step through an install (windows side only)

Needless to say I'm still figuring out how it all works.

Sunday, October 26, 2008

open source


FSOSS was great. For me Friday was it. The two workshops I attended were very high calibre, two dynamic presenters and content that was itself an eclectic mix of dumming down to meet the audience needs and then the real thing once people were 'getting it'.

Friday, October 17, 2008

Wubi port project


This is my Fedora project. Port Wubi to Fedora.
I've been dragging my feet on this soooo, time to turn the page. I just got the source code for the new Wubi/lupin builds. Pure python. That surprised me - I had expected it to be a combination of C++ and python (with python replacing the earlier NSIS stuff) - even though I was told it was being written in python. The new code will be part of the Ubuntu 9.04 Just to make it very clear - I have no part in the Wubi Ubuntu installer project.

I will be using the new code - I am very grateful to Agostino Russo, the head of the Wubi project for giving me access to this code, I feel privileged. So far, I'v just set up my XP environment - I guess I didn't really need those compilers anyway :) well they are still useful for compiling the older Wubi which I will also need to learn from.

Anyway, thanks alot to the Wubi team!

SPR720 Lab, RPM creation sqlite exit 0

Finally, couldn't believe the solution to my problem. error: Installed (but unpackaged) file(s) found:

I reread the Fedora book package chapter and the Fedora site info
PackageMaintainers/CreatingPackageHowTo

ONly thing i didn't do was run out and the Maximum RPM book (I may do that yet).

I simply copied/pasted the list of files under the error message under the %files section and it built. This was a very useful exercise. I want to look for something I can package and try to submit to Fedora (will need to find something I already know so it won't take too much time). It would be good to go through the whole process even ifit is not finally accepted the insight on how this works would be very useful for the future.

I guess the spec file is redundant since it will be in the other rpms anyway.

sqlite-3.6.3-1.fc9.i386.rpm


sqlite-3.6.3-1.fc9.src.rpm

Sunday, October 5, 2008

gettin back on track


Ohhh too many distractions - got unix in a nusthell at the second hand shop 49cents
and a bunch of stuff out to the way so now ill get down to work

Just replaced the hard drive in my vaio and for 60 bucks it is working again (great I even have some screws left over) put a 20 G Ubuntu partition, the only thing I had handy, and put the RPM package on so I can work on RPM a bit before our guest speaker (Andrew Overholdt - he coming at Red Hat) tomorrow. I'll try to get Fedora on it too before his talk so I won't miss anything.

My Eee Pc is back running Xandros - I'll look at their wiki and figure out what I need to do (wifi drivers) to get Fedora going on it - later. Going through Chapter 5 on the Fedora Linux (tyler, 07) book now. I'll run a bunch of rpm commands trying out the combination of package queries and output options and put the results on a wiki here . Once I get the idea I'll start on the lab for the scripting class... more later

The presentation by the Red Hat package maintainers was great. I'm glad I am doing this after their presentation. There is a lot to it but they went through the whole process and explained everything as they went. Not just that though, they went through an example of submitting a package for review using a real world example (the package author was in the room) and even using the koji build system. There was a lot of information for a one hour session: my own focus was on their detailed analysis of the .spec file (that specifies the packaging build process). I regret that I didn't take advantage of the question period though, should have had some dumb question on hand since they were there and willing. Maybe, should have asked "what happens if you are building for both Solaris and Linux?", but I was too timid - well next time!

OK, so I am building and RPM for sqlite. That is the cool little database that doesn't require a server in the traditional way that mysql does. I pick this program for all my examples as I want to become really familiar with it - this one is important.

All the gory details are on my RpmLearn page

Here is the description of the source from the README:
The README: This package contains:

* the SQLite library amalgamation (single file) source code distribution,
* the shell.c file used to build the sqlite3 shell too, and
* the sqlite3.h and sqlite3ext.h header files required to link programs
and sqlite extensions against the installed libary.
* autoconf/automake installation infrastucture.

Following the advice from the PackageMaintainers pagepage, I installed sqlite from source on my Fedora laptop. I put all the guck here so I don't lose it

you know the routine: ./configure, make, make install

gonna hang this on matrix soon as i have time
http://devzone.zend.com/node/view/id/641 that is a little php/mysql tutorial that works



Now back to making an RPM. Specfile sqlite.spec here.

Sunday, September 28, 2008

installfest experience



All kinds of people showed up. I met half the bio-informatics people from Seneca's grad program. I was wearing a 'Linux Installer' name tag, but in reality, I was there to learn from the experts. They didn't let me down. I learned a lot - just scratching the surface though.

http://www.flickr.com/photos/25612516@N08/


.
.
.
.
.
.
.
.

SPR720 Lab, compile and run SQLite


Our Lab involved getting the source for something interesting and compiling and using it. It was fun to do: there were no dependencies issues and all went well. I had trouble trying to talk sql with it though but the prof had it working properly in a minute. I'll experiment with it a bit as I want to put some practice web db stuff on my matrix account. Good fun!

sqlite> CREATE TABLE `symbols` (
...> `id` int(11),
...> `country` varchar(255) NOT NULL default '',
...> `animal` varchar(255) NOT NULL default '');
sqlite> INSERT INTO `symbols` VALUES (1, 'America', 'eagle');
sqlite> INSERT INTO `symbols` VALUES (2, 'China', 'dragon');
sqlite> INSERT INTO `symbols` VALUES (3, 'England', 'lion');
sqlite> INSERT INTO `symbols` VALUES (4, 'India', 'tiger');
sqlite> INSERT INTO `symbols` VALUES (5, 'Australia', 'kangaroo');
sqlite> INSERT INTO `symbols` VALUES (6, 'Norway', 'elk');
sqlite> select * from symbole;
SQL error: no such table: symbole
sqlite> select * from symbols;
1|America|eagle
2|China|dragon
3|England|lion
4|India|tiger
5|Australia|kangaroo
6|Norway|elk
sqlite>


Figure out how to change these connect and select statements to sqlite this will be useful
// open connection
$connection = mysql_connect($host, $user, $pass) or die ("Unable to connect!");
// select database
mysql_select_db($db) or die ("Unable to select database!");
// create query
$query = "SELECT * FROM $symbols";
// execute query
$result = mysql_query($query) or die ("Error in query: $query. ".mysql_error());

Monday, September 15, 2008

UPDATE --- Major Project Selection

This last Saturday, I had the opportunity to see many Linux instalations taking place at Seneca College's Linux Installfest. Due to the fact that Seneca is well connected with the Fedora Linux distribution community, I had expected to see a lot of Fedora installations.

What I saw was an overall preference for Ubuntu Linux. Most installations seem to be dual boot Ubuntu and Windows Vista systems. I am new to Ubuntu (I use Fedora) but Ihave noticed that Ubuntu is being discussed more than other distributions on forums I follow: techsoup.org for one. I am curious about why this distribution seems to have a wider following than other distros.

train is coming into the station so i have to go.

I plan to mention Wubi eventually...

There is a negative comment/observation below which I will leave in tact to remind me how difficult I think this is.

I'll add a few \ns later:

Wubi is the brainchild of Agostino Russo. From what I understand, Ag or xivulon, had monitored the experiences of people (normal people, not programmers) with regards to attempting to switch from licenced OSes to free systems like Linux. He noticed that people were struggling with the fundamentals like running fdisk and everything todo with partitions. So he automated the process. As he puts it (actually mmy paraphrase), 'in Windows, you click on a windows executable (say wubi-installer.exe) answer a couple of questions then go get yourself a coffee and when you get back it will be done - you reboot and you have a dual-bootsystem. Now that is cool. It should come as no surprise that Linux systems are more and more into the mainstream. It is my belief that Linux distributers must adopt a new approach, a new mindset wwith regards to attracting users. Users should be considered to be customers, clients whatever you want to call them but treat them like they were actually paying customers. I think that the whole Linux distribution system is broken. Because Linux is free, everybody, form software developers to distribution executives expect their users, their customers! to understand things that only system administrators should have to know. The other OSes don't do that - they never have done it. Ask any MS Windows user what she thinks of fdisk and you will get nothing but a funny look - and rightly so because WHO CARES!!! SysAdmins yse, users no. It is the philisophy driving things like Wubi that is will fix the broken Linux distribution model More later.




Looking, but haven't got there yet. After briefly thinking about the 'Wubi - port to Fedora' I have definitely ruled that one out. My thinking is that one needs solid grounding in at least on of those OS to start with - I don't have that.

Thought on it though: Great idea! I think that it is things like Wubi that will enable Windows users to make the leap...

Noticed that somewhere on some forum ut was called 'Linux for cowards'; hey I love it.

Looking at creating 'Windows migration tool' now (on fedorproject site summer of code 2008).
You see I do think migration is a very important topic.

Friday, September 12, 2008

Wednesday, September 10, 2008

LPT370 Assignment 1


software patents

the good, the bad, and the ... (already been said)

Using technology to engineer social change?? How on earth could technology be used to affect the human psyche.

"While an algorithm standing alone can't be patented, if it can be demonstrated that that algorithm can achieve some commercial aim in its application in software, well ... that's patentable. Well duh" - Mark Webbink - Truth Happens

look and listen here:
http://truthhappens.redhatmagazine.com/2007/10/19/mark-webbink-on-software-patents/

Sure seems like a complicated subject: all these genius lawyer types butting heads over things I can can barely grasp at all: I even get stuck on the algorithms.

But seriously, isn't the whole argument slightly inane? Nobody seems to be arguing that patents are wrong the whole peer-to-patent thing is fighting to improve the process of patenting. So, patents are good, patent protection is good and everything for sale in the marketplace is good. The bad - well that must be those folks we don't agree with, the ones who patented that single click mouse thing. Reminds me of the old saw about everything being for sale - it is the 'price' that is to be negotiated.

I gather, from listening to Webbink, that packaging (the algorithm) has a lot to do with it being patentable hence sale-able. It is software that has to demonstrate the ability to satisfy a commercial aim to be patentable.

That is all well and good but, it is the standalone algorithm - the art - in the Art of Computer Programming (see Knuth, 1968), that should be patentable - make them prints for the wall! Yes, make it art we can use but patent it for what it really is - and yes, lets sell it too.

Oh and there is a nice message in the preface of the book (the image at the top of the post).







What is with bill C-61



Was that visual basic © running on the gov't website? So I need VB to get a government job?

I usually leave this type of stuff to the thinkers but since I have to...
I'll spend a few minutes researching C-61 then a few more lying about it.

Oh boy! This looks ominous - what have they got to do with the Internet anyway. I thought the whole idea behind it was the global village, it's sans frontiers - none of their business.

From the federal website.

C-61 SUMMARY
This enactment amends the Copyright Act in order to
(a) update the rights and protections of copyright owners to better address the Internet, in line with international standards;
(b) clarify the liability of Internet service providers;
(c) permit certain uses for educational and research purposes of Internet and other digital technologies to facilitate technology-enhanced learning, inter-library loans, the delivery of educational material and access to publicly available material on the Internet;
(d) permit certain uses of copyright material for private purposes; and
(e) amend provisions of the Act relating to photographs to give photographers the same rights as other creators.




If they would just stop mentioning networks and the Internet I could probably overlook this.
I prefer not to get involved but sometimes - it's hold my coat and give me a rock -
I recall, in an earlier lifetime, a lady I knew was stopped boarding the DC-9 at Timmins airport as she was carrying a large bread knife (might have been in her purse she never got on the plane), well, Kiki Lapier was on her way to Ottawa to straighten out one PET - bare with me if this seems slightly off topic, it is a tale of righteous indignation - you see Kiki (she was a lady who never took authority well) had a 4*8 sign on the roof of her house that simply said 'seat belt law fuddle duddle', she had had enough.

Oh, sorry to go off topic. Anyway, in honor of Kiki and her personal crusade, until the C-61 issue is resolved, though I may buckle up I vow to silently utter that oath - fuddle duddle - whenever I do. And, more than that comrades ... at least in my mind, I'll see you on the barricades.

Monday, September 8, 2008

SPR720 Command Lab


Linux commands

and comfort zones: some thoughts and a comment on the lab
dos 6 (©) was pretty cool: I got used to it and I liked the simplicity (if only I could have connected the network with those borrowed token ring cards). Still, it was a great environment to try out that Bor£and compiler though.

I've always found the Unix directory
structure confusing. Even after
getting fairly comfortable as a user
the file system remained a mystery.
Thankfully, the whole user experience
is getting easier as Linux
distributions evolve.

In Lab #0 we took a look at what is
in all those 'bin' directories.

What is with all these bin directories
anyway? That is one of the things that
always made me uncomfortable:
especially when I was trying to find
something. To me, the absolute worst
feeling was downloading installing something and having no idea where all that stuff has gone or even what exactly I had installed (doc filesuseful tools...).
Hey, usually everything worked but I was clueless and usually had to 'get back to work' so no time to spend exploring.

Coming from a dos background, I wanted and expected to know where things were. Put ActivePerl on a Windows machine and you get one directory, say C:\>perl and all that neat stuff is in there; dead easy to browse the docs, look at the samples - I guess it didn't make Perl any easier though.

About the Lab: we took a look at what was in those bin directories, each file (maybe not the thousands in /usr/bin) or at leas some of them; we found 25 files we didn't know and did a little research on them - man pages, info, --help etc. Then think about what was going on withe the different dierectories. There is obviously rhyme and reason for the structure, the multiple bin directories.

First of all, the Lab was deceptively simple. It took a lot longer than I tought it would. I spent time looking at all those /usr/bin/mysql* files that I knew (from my C:\>mysql directory) were around - I will have to go back there and do that some more. I discovered lots of little thing that were interesting. I'd started out with the intention of looking at all the two letter commands as they tend to be the heart and soul of Unix systems, powerful little things that the gurus pipe things to and all that. I discovered a whole lot of z* commands that deal with compressed files - those are not in my comfort zone yet.

Anyway, I made some minor conclusions (temporary and ready to be rewritten) about the reason for the multiple bin directories. Part of it has to be historic - that is the way we've always done it - but there is definitely a seperation of system stuff from user stuff (who is the system and who is the user...).

Out of time...

SYA710 Lab 0



Hardware based data recovery is definitely a plus in a volatile environment such as the Seneca Lab.
Without even considering the possibility of malicious attacks - rather thinking about accidental or inadvertent mishaps - this Lab has people who are new to the system and need to experiment to get
their work done. Accidents do happen: people delete files and sometimes something gets trashed that should have been left alone - everybody who spends a lot of time working on computers has done it at some time. So in the Seneca Lab it makes good sense to take this proactive step.

The same surely cannot be a good thing when applied to workstations that are used by people doing serious work that involves the need to set up special environments and have job-specific tools on their computer. I would have to think that traditional timely backups would provide a sufficient protection for recovery.

This Lab was useful in that it actually demonstrated that the recovery does work. If somebody told me that it worked but I didn't prove it I would likely have a slight doubt in the back of my mind - does it really work; maybe it works sometimes...

The one problem that I had with the lab (the assignment) was that my scp command doesn't seem to be doing what i thought it would. The command (scp lab00.txt xxxx@xxxxxx.senecac.on.ca )
executes without any message - no challenge for password - but doesn't copy the file. I am executing the command from home and my network connection is working as i could mail the file to myself. I just tried scp xxxx@xxxxxx.senecac.on.ca: laboo.txt it connects, authenticates but gives the message 'scp: .: not a regular file'

LPT370 Assignment 0


Temeraire on its way to the breakers

Time_to_do_things_differently

LPT370 assignment 0.

XV - Cool image manipulation tool for X Windows. Shareware (not open source but the source code is available).
http://trilon.com

This tool has been around longer than the Web. Its author even calls it a 90s type image editor. I think that was when I first used it. The really interesting thing about this tool is that I can do the same things with it that I do with Photoshop, PaintShop and other 'humungus' tools. But there again I only do really basic things so don't throw away the big tools just keep this one handy for lightweight needs.

I found this tool was mentioned in a tutorial about streaming images a byte at a time and the demo showed a GIF image with a small colour table that was easy to change (and easy to see the effects due to the limited number of colours).

The image (from the tutorial mentioned above) has only a small number of colours as can be seen in the editors view of the vx colour editor. By simply clicking a button (or manually text editing the colour map itself) xv lets you manipulate the colours of the image. For example, in the image below the colour table was rewritten by clicking the 'grey' button.

The really interesting thing about tools like this is that they can be used in conjunction with other tools programatically. When I use Photoshop I don't even think about what is going on
behind the UI. With a tool like this the whole editing process takes on a different feel. Tools like this invite the user to think about what is going on. Sure, the algorithms
would be the same whether one edits with Photoshop or xv but xv lets me feel 'in control': nice illusion :)










SL

Client




A brief description of the Second Life client. First, although it is usually a surprise to SL users, not everybody knows about it.

A couple of difinitions:

From the website http://secondlife.com we get "Second Life is an online, 3D virtual world imagined and created by its residents" (residents == users?).

Googling 'what is second life' gives an assortment saying, typically: "Second Life is an online 3D interactive virtual reality program which resembles console video games such as Final Fantasy, but is almost entirely built and influenced by the people who use it. 'In world', you use a highly-customisable avatar to explore the thousands of places to visit, build and interact with the world and other users."

SL has a marketplace, institutions, movements pretty much like the real world as people map their experience from one life to the other.

The SL client or viewer is works like a web browser but everything is 3D. One idea is that it what the rest of the web browsers will look and feel like in a couple of years. Through the viewer communication with other 'residents' in chat like fashion or using voice. The viewer has functionality to search, shop and generally do things one does on the Web with a browser in real life (RL) except it is within the virtual world. All of this happens in a highly developed 3D environment.


The SL client riquires a reasonably powerful computer due to the graphics. The source code for the viewer is apparently now in the public domain.

i'm a rookie but this is ... well what it is


http://www.youtube.com/v/m3TD1hty0Aw&hl=en&fs
from the author:
an introduction to Second Life at the Summer institute on teaching and technology at University of California Davis - joint presentation and workshop by Milmon, Bernadette and Sandy (Harrison Held, HVX Silverstar and Agnesa Capalini in SL).
if watched in the 'high quality' the annotations do not show (bug in youtube's annotations feature... but still a very cool feature.
you can add callouts, labels and tag sections of the video. There is also an option to add a hyperlink to the rollovers, but if you follow the hyperlinks you will lose your spot in the movie.. it is not yet that intelligent
but cool that you can add a link to another youtube movie at the end of your own video..
only links to youtube vids allowed.
There's an avatar taking a look around a music shop... and he's sporting a Tux T... the official mascot of the Linux kernel... because it's waaaaay cool.
Tux icon on avatar's T

Thursday, September 4, 2008

begin again - this is about me

I think I could if I only knew how to begin.
Is that a quote from Lewis Carroll?
If you know let me know.

something about myself
16.5 high school credits, electronic engineering techniques diploma (seneca'81), ba english (york'94), ba computer Science (york'99), some open university grad credits, ged'06



I like old artifacts, cars in fact, but all things related - it it the little things in them, the period guages, the steel with hand made compound high crown curves hand formed, leather and wood - oh and the wind in the hair.

href="http://slowtek.wordpress.com/2008/09/03/hi this is where my personal stuff is (i am learning wordpress)



i'll finish this later ...