[ICO]NameLast modifiedSizeDescription

[PARENTDIR]Parent Directory  -  

This website is under control of the stupid content tracker (git).

I made ~lam/public_html a git repository on Sunday June 21, 2020

You could clone this on triton with:
git clone ~lam/public_html
You can examine the log on triton with:
cd ~lam/public_html
git log
You can use git status|diff and other commands that require only read access.

At the current time the RCS archives remain because they have some information on changes made before Sunday June 21, 2020. I can't remember if I had checked whether triton had git installed on it when I discovered it did not have rcs. I am over 65 and learned revision control long before git came around.

Why now?

O.K. I only got around to getting this website is under control of git Sunday June 21, 2020 but it's a tiny site using less than 1/2 MB of space including the git repository. I first used rcs even though I had to build rcs on triton to use it. The build was partly and maybe even mostly to test my ability to build an application on triton but it was also because rcs is often the first thing I look for when I want to put a file under revision control. When I get on a new Linux/Unix system I use it to archive the .bashrc .bash_aliases and any other files I am going to change. If I am administrator of a system I do the same thing for /etc files I change to control the operational environment.

Once you want to control multiple files git is a much better revision control system for a large variety of reasons. The main reason is the commit which can be for anything from changes addition or deletion of a single object to a change affecting every object currently under it's control.

I still use rcs for a large part because I used it decades before git even existed. Although git has been around more than a decade, rcs was initially released over 38 years ago when I was getting my Computer Science degree in my earliest days of coding. As of today, 1/3 of the time git has been around is after I retired.

Something to do

There are a few reasons I am getting around to this today. One is that one of my main activities while practicing social distancing because of the COVID-19 pandemic is playing with web technology. A related reason is me getting around to a "Set up a GitHub Pages website" task I had added to my ToDo April 29, 2015 a few months before I retired from Arctic Region Supercomputing Center (ARSC) because it ceased to exist. While worked for ARSC I worked on projects that used git but none of those were uploaded to GitHub let alone any "public" repository. For most of the time I worked with git, a repository was a local path (often on a shared filesystem mounted locally) or a system:path combination. In both cases it was the same argument that would work with rsync. I created my GitHub account as part of The Data Scientist’s Toolbox course from Johns Hopkins University which is a MOOC (Massive Open Online Course) earlier in 2015.

So this site is one of many I decided to use git for revision control but it was a little different in that I had already implemented a clone of the site. The clone of the site is updated daily by a rsync job that uses this site as the source. The clone also creates a backup using tar that is updated during a daily check if the content has changed. This is the source used in the CloudInit directives when a new LAM AWS EC2 instance is launched to take over hosting of the site.

The .htaccess file that has some Apache configuration directives for the lam.blinkenshell.org site was already within the repo because of the way web pages at BlinkenShell works. I decided to move the Apache2 configuration file for the clone into the repo as well. While I was at it I moved the blinkenshell_archive_rebuild and blinkenshell_aws_rsync scripts into the archive and updated some of the comments.

I then decided to create this page.

upThis page up was last updated Monday, June 22, 2020 @ 11:00 PM (Alaska Time)