life of a gizmo

Wings for my business and why google is the BEST

2010-08-19T07:00:51Z

My Web Design Business

I just launched my web design business based in Lethbridge. This is something I’ve wanted to do since high school.

There’s two types of people in the world: dreamers and doers. I’ve been (and still am to an extent) a dreamer. Dreamers are people with ideas. They’ll tell you about these ideas until the cows come home (I wonder where that expression comes from. Cows don’t graze on their own, but I heard goats do! Maybe it should be “until the goats come home”). Often these ideas are great and if executed properly could turn out to be BIG. However, that’s just it, they don’t execute and let their ideas rot. I’m moving away from being a dreamer to being a doer. Doers execute in rapid succession. While dreamers talk about the awesome they’re going to make, doers are already doing it (and maybe failing and trying again!).

I’ve dreamt of having a web design business since high school. I’ve even got mockups to prove it! I took forever to pick and name then didn’t like it. Took forever to pick another one, registered a domain name. It went stale and I didn’t like that one too. Rinse and repeat. That’s until now. I pray to God that I start becoming more of a doer as dreaming alone leaves you immobile.

Why Google is the BEST

Google has something called Google Webmaster Tools. As the name suggests, they are meant to help webmasters help google discover their websites. I used it, it worked!

That same google page, kindly enough, mentioned that Bing and Yahoo! also have their version of webmaster tools. I tried both of them with nothing but problems! That’s why google is the best. I didn’t think the Google tools were special because they just worked, as expected. This was until I tried Bing and Yahoo! where the tools failed miserably. Especially Bing’s.

CSRF verification failed. Request aborted.

2010-08-13T06:23:26Z

You may be doing your daily business, as I was, then get an email from a client saying there’s a problem with the (django) site you created. You think to yourself, hmm, that’s odd, nothing has changed in that since since X years.

You check out the site and try to login only to discover this nasty error:

FORBIDDEN
CSRF verification failed. Request aborted.

This is most likely caused by an intentional or unintentional upgrade to Django 1.2.x. For me this upgrade was unintentional. My host is dreamhost and they decided to upgrade their django version to 1.2.1. Once this happened I started getting that error out of the blue! If you did upgrade to django 1.2.x intentionally, you need to make sure you read the release notes and pay close attention to the backward incompatible changes. You will probably have to make changes to your code.

I already had a virtual-env with the old version of django installed, so all I had to do was change these two lines of code in my ./dispatch.fcgi.

Before the change:

sys.path += ['/home/myuser/django_src']
sys.path += ['/home/myuser/beta_django_projects']

After the change:

sys.path.insert(1, '/home/myuser/django_src')
sys.path.insert(1, '/home/myuser/beta_django_projects')

This sets the your django_src path to higher priority in your PYTHONPATH which results in your django version being used over the one available at dreamhost.

These instructions are very tailored to my dreamhost issue. If you something is not clear or this doesn’t solve your CSRF problem, just leave a comment and I’ll try to help you out.

Lethbridge Roofing and Insulation

2010-08-04T19:43:03Z

Pleasure to Work With You

Darren, the general manager at BHL, was essentially my first web design client. He’s an amazing guy to work with. No hassle and no pain. I do the work, get feedback, implement it, and he’s happy . Have you heard about the client from hell? Well, Darren, is the exact opposite!

Lethbridge Roofing and Insulation

Today, we just launched the website for his company, BHL. BHL provides a number of services including roofing, insulation, sound proofing, fireproofing, construction for the Lethbridge area.

Highly Recommended

I highly recommend Darren and BHL for Lethbridge roofing and the other services they provide. Truly a great person to work with.

Firsthand Experience with the Personal Software Process, part 2

2010-07-30T08:02:45Z

This is a series of three posts on my experience with the PSP.
PSP stands for Personal Software Process. For an introduction, see the wikipedia page on PSP.

Part 1 was posted recently. Enjoy part 2:

PSP Levels

PSP℠ has been developed with training in mind. It starts you off with PSP 0 which is essentially your existing software process but adds measurements and through a series of six steps exposes the student to all of the PSP processes and measures. Below I will describe each step in more detail and explain the benefits of each.

PSP 0

Description

As mentioned above, this level attempts to establish a baseline for your existing development practices. It does, however, “force” you to split your development into phases: planning, development (design, code, compile, test) and postmortem. It also teaches you how to record development time and defects introduced. The only ‘planning’ aspect of it involves predicting how long each phase will take you. As you approach PSP 2.1, this data will show you how your predictions are better with PSP than without it–or not!

Takeaway

It’s very tempting to start developing a project (especially when it starts out small) by opening an editor and coding. Compile. Fix errors. Repeat until good enough. From personal experience, this leads to a lot of minuscule bugs because no attention to detail has been paid. It’s easier to let the compiler find the bugs for you! Using PSP however, you’re forced to plan, design, code, compile, and test. When I started using PSP 0, I noticed I was a lot more careful even with the planning phase. I gave some thought to what the design and coding will encompass in order to estimate how long each phase would take me. Normally these steps would have been performed implicitly, if at all. In the design phase, I started giving more importance to defects and payed close attention to design completeness as I didn’t want defects to appear on the defect logs! Something magical happens when you actually track even your defects that don’t make it into the “field.”

In the postmortem phase, it was definitely eye opening to find that I spent 49 minutes in testing! That’s 46 more minutes than I predicted!

PSP 0.1

Description

This level adds program size measurements, process improvement proposal, and a coding standard. The process improvement proposal is “a form that engineers use to record the process problems they encounter as well as their ideas for addressing them.” [4]

Takeaway

This level was a little more frustrating than the first. I think the challenge of doing a self-study of PSP is that instructions tend to lack a great deal (from the online materials). I initially didn’t understand (and still might not) the difference between base, deleted, modified, added, reused, and added and modified program size aspects. The especially annoying bit here is the reused category. In my opinion, since, I’m not designing libraries, my code is never 100% reusable like a library is meant to be.

In addition, the process improvement proposal (PIP) form, I found, was not very useful as again, I wasn’t sure what parts of the process I could question or what I could do about it to begin with. In hindsight, I’m glad I stuck through with most of the process suggestions as I now have a better understanding and appreciation of some of them.

On a more positive note, the coding standard was useful. As developers, we often have a personal coding style but, unfortunately, if it’s not documented we often end up straying from it.

PSP 1.0

Description

This level was exciting. You will finally reap some benefits. “In this level, engineers are introduced to the PROBE method, which uses historical data to estimate size and determine the accuracy of the estimate. PROBE is a regression-based size-estimating method developed specifically for PSP.”[4]

In addition to previous data, this level asks you to predict program size by breaking down the design into parts added and reused.

Takeaway

At this phase, you will get a lot more involved in the design as you are asked to list out the parts (functions, classes, modules) you’ll need to design your program. The notes on this level tell you to create a conceptual design in the planning phase in order to create somewhat of a parts inventory to allow you to predict how long each part and thus the whole program will take. Up to this point I was using a word processor to keep track of all my data which made the PROBE calculations a little cumbersome. These in the later phases lead to me dreading the planning phase the most! Luckily though, there’s a tool that makes all of this really easy, so use it: Process Dashboard.

PSP 1.1

Description

“This level adds resource and schedule estimating and earned-value tracking. Engineers often have trouble tracking their work because they do tasks in an order different from their plan. Earned-value tracking allows them to weight the relative importance of each task and to judge their progress as they finish some tasks early and others late.” [4]

Takeaway

The additions in this level seem really promising. It’s quite easy to get distracted by different projects at different times while letting other priorities slip. However, with the schedule and task planning you can keep track of your progress and your projected finish date. This allows you to renegotiate a deadline if you’re sure you will miss it. Having said that, the task planning and schedule planning template are meant to be used in project that will take more than one day. I attempted to use this in my 5th project, however, it didn’t prove too useful as it did not take more than one day. Interestingly enough, in the project that I should have used these tools (Project 6), I didn’t, since I anticipated to spend about 8 hours on it instead of the 45 hours the PROBE methods suggested! (Hint: I spent about 50 hours on Project 6!)

PSP 2.0

Description

“This level introduces design and code reviews, as well as quality measurement and evaluation. Using defect data from their earlier exercises, engineers also develop personal design and code review checklists.” [4]

PSP 2.0 does come with prebuilt design and code review checklists.

Takeaway

When completing the checklists you’re not supposed to work on more than one item at a time. For example, if you’re checking for include statements, that’s all you’re doing. These checklists require a lot of discipline to not rush and get them over with. Perhaps, if feasible, it’s a good idea to postpone these reviews for some time (be it hours or days), in order not to speed through them and miss critical defects. From my personal experience, although I have not followed these to the dot (because some didn’t apply and some seemed too burdensome) they have proved very useful. A much better practice, which I shall do in the future, is to modify these checklists to suite your needs by examining the time of defects you’ve produced in the past.

PSP 2.1

Description

This level breaks down the design phase into functional specification, operational specification, logic specifications, state specification. The functional specification is essentially a UML diagram with method specifications. An operational specification can be compared to a use-case diagram. The logic specification is, as the name suggests, the logic of an item (e.g. method) and finally a state specification is a state machine.

Takeaway

I used PSP 2.1 for Project 6 which was essentially maintenance work on my textbook website. Instead of following these different specifications, I decided that a UML diagram representing the changes I intend to make to the software would be sufficient. I also added documentation to the classes and functions in order to cover functional specifications. As the use cases for my website are well defined as it’s already public, I didn’t feel a need for the operational specification or use-case diagrams. Likewise with the task at hand, a logic specification wasn’t needed as I was replacing a module with another one and not fixing any existing logical defects. And finally, a state chart diagram seemed too cumbersome to even start–so I didn’t! These pain points of the PSP will be reflected in my own process which I will introduce below.

Part 3 will be posted shortly. It basically describes some changes I intend to make to PSP to make it work for me. PSP is meant to be personalized and cusotmized, hence the name, Personal Software Process.

References

[1] http://en.wikipedia.org/wiki/Personal_Software_Process
[2] http://www.methodsandtools.com/archive/archive.php?id=60
[3] http://www.cs.uleth.ca/~rice/cs3720/papers/Cannon1999_PSPinpractice.pdf
[4] http://www.cs.uleth.ca/~rice/cs3720/papers/Humphrey1996-ResultsPSP.pdf

Firsthand Experience with the Personal Software Process, part 1

2010-07-30T08:00:17Z

This is a series of three posts on my experience with the PSP.

PSP stands for Personal Software Process. For an introduction, see the wikipedia page on PSP. This is part 1. Part 2 can be found here.

What is the Personal Software Process℠?

The Personal Software Process (PSP℠) was developed by Watts S. Humphrey from the Software Engineering Institution of Carnegie Mellon University. PSP℠ is a highly disciplined software development process. However that’s not all–it’s even described as the process that helps you develop software processes and not necessary a software process in itself. PSP℠ focuses on strict measurement of software development activities starting from planning, designing, coding, compiling, testing, and postmortem.

Why should you care?

In my Software Engineering class, through a number of papers (see References for a list), I was introduced to the Personal Software Process℠. These papers made some very seductive claims on improved completion estimates and software quality and they got my attention. Here are PSP success stories:

At Advanced Information Services (AIS), before PSP training, schedule estimating error averaged 394 percent; afterward PSP training the average was -10.4 percent. That is, projects were being completed even before their estimated deadline.[4]
Also at AIS, the software for a PSP trained team had 0.76 defects per 1,000 lines of code before PSP training and 0.17 defects per 1,000 lines of code after training, a 78 percent improvement. [4]
In addition to the above, AIS compared the results of project with PSP trained engineers to those with engineers not trained in PSP. Here are the results [4]:
- Project C (completed by non-PSP trained engineers) was scheduled for two months, but took five to reach the acceptance test stage.
- Likewise Project D (again, completed by non-PSP trained engineers) was scheduled for 10 months, but took 19.
- Project B, however, was scheduled for seven months, but took five. This was completed by PSP trained engineers.
At Motorola, of eighteen projects completed by PSP-trained engineers, only one defect had been found after release.[2]
At Union Switch and Signal, PSP-trained engineers completed five projects and no defects had been found after release.
And finally, in my PSP training Project 1 consisted of a trivial 31 lines of code (LOC) which took 2.38 hours to complete but my schedule estimation error 26.55 percent. Project 6, on the other hand was a much more complex maintenance project with 247 LOC which took 50.42 hours to complete and I only had 10.36 percent schedule estimation error. (In reality if I had estimated Project 6 without the PSP tools like I estimated Project 1, I would have estimated development time to be about 8 hours which would have resulted in a 92.85 percent in estimation error!)

Further, you might wonder, while those numbers look really good, why are they so important. When it comes to schedule estimation, it’s important to get it right because you likely will be paying people (or are getting paid) to work on the software. These costs are covered by the client of the software (otherwise you’re not making a profit). If you underestimate how long it will take to develop a particular product (be it custom or packaged software), you’re bound to charge less than feasible prices and end up incurring a loss. This can lead your business (or employer) into trouble.

Likewise, quality is very important. Removing defects gets more expensive as you progress in the development of your software (e.g. a defect that might cost one dollar to fix in the requirements phase may cost $10 during design, $100 during coding, $1,000 in test and $10,000 in the field.)[2] Again, if you reduce the number of defects that make it to the latter phases of a project you are contributing to your company’s (or your employer’s) bottom line.

In a nutshell, that’s why schedule estimation and product quality are important in a software project. That is, without even considering the marketing benefits reaped from a quality product and good estimation.

How do I start?

If you’ve come to the conclusion, as I did, given these potential benefits it is worth trying PSP out, I recommend you start with the Self-Study PSP material you can download from the SEI website.

References

That was part 1. I’d love to hear your opinion on this especially if you’ve tried PSP or a “competing” development process like SCRUM, etc. Until part 2!

Refreshing: Button Feedback

2010-07-09T21:16:02Z

I came across printfriendly.com only to realize it’s user friendly too.

One of the things that has bugged me but not enough to notice it, has been lack of feedback on a form (like a search form). You enter the text you wish to search for, hit Enter on the keyboard and you see nothing until you look at the browser loading bar to notice that “it’s doing something.”

Normal button state. Dark gray background color.

Button feedback--flashes blue

What printfriendly has done though, is added feedback to the form. When you hit Enter, the “print preview” button flashes letting you know that, yes sir, your request went through! This type of feedback is common for mouse clicks and I’m glad someone has thought of people who just hit Enter!

deploying websites on virtualenvs with deploy_pysite.sh

2010-07-08T00:18:10Z

The Story

I had some work to do but instead found a perfect distraction. I decided to upgrade my textbook website to Django 1.2.x which is an updated version of the python framework I used to build it.

Unlike in the past, where such a task would have not phased me and which I would have attempted to do on the live site, I decided to setup a beta site that runs along the main GizmoBooks website as a sandbox so to speak.

I started upgrading and it turned out there were many changes that needed to be made. I came across this very helpful guide on intalling python 2.6, virtualenv, and virtualenvwrapper. After a few attempts and tweaks it worked and beautifully too! Not only did I have python 2.6 installed but I also had virtualenv and virtualenvwrapper which I until I had it, I had no idea what I was missing. It’s quite brilliant because it allows you to setup virtualenvironments like this *snaps fingers*.

So far so good, I had my virtual environment setup but then came time to find all the different python PIP packages and other software that my website needed. This was okay for a bit, but then came time to install xapian-core, xapian-binding, blah blah blah. I quickly got tired of all this.

The Script

So, I did what all lazy programmers do: I wrote a script about it! The script is called deploy_pysite.sh and can be found at GitHub.

You see, the beauty of virtualenv is that it allows you to quickly create a virtual environment from scratch. The problem, of course, is that (often) the virtualenv is empty and you need to go install all the packages to deploy your website. This is what the scripts aims to simplify.

Script Guide

Follow these steps to run it:

If you’re not using virtualenvwrapper, set VIRTUAL_ENV environment variable to your absolute virtual environment path. If you are using virtualenvwrapper, just “workon” the virtual env you wish to install the packages in.
Add to, or remove from nonpip_packages and pip_packages arrays in SETTINGS section of deploy_pysite.sh source code. Use the formats specified in comments.
(Optional) The default directory where software is downloaded and built is $HOME/downloads. This is set in dg_downloads_dir in deploy_pysite.sh SETTINGS section. Feel free to modify it.
Execute the script.

That’s pretty much all there is to it.

Not surprisingly, it proved usefulx100 immediatly. You see, I was following another guide from andrew.io on installing Django with virtualenv on dreamhost and low and behold the guide said to “[i]nstall any packages you may need for your project.” ./deploy_pysite.sh was all I had to say at that point .

Script Source

I’ll post the source here, but it’s probably best to just grab it from GitHub.

#!/bin/bash
 
# ensure that exceptions are treated as such and program exists!
set -e
 
# =================================================
# = deploy_pysite version 0.1.0
# = 
# = Created by Gezim Hoxha
# = License: GPL 2
# =
# = If you're using virtualenv, this script will
# =     install PIP packages and even build 
# =     build packages from source then install
# =     them into your virtual environment.
# = See SETTINGS section for instructions.
# =
# = July, 2010
# =
# = TO DO
# = - 
# =================================================
# = REQUIRES
# = bash >=3.1
# = curl
# = pip
# =================================================
 
usage="
USAGE: \n
Modify nonpip_packages and pip_packages according to your needs. These can be found in the source
of this script in SETTINGS section.\n
Once SETTINGS have been modified, activate your virtual environment, then call this script. \n
If you're not using virtualenvwrapper, set VIRTUAL_ENV environment variable 
to your virtual environemt directory. \n
\n
Also, you may want to ensure that the download dir is set to where you want it in SETTINGS section.
"
 
# ensure we're in a virtual environment
 
if [[ "$VIRTUAL_ENV" = "" ]]
then
    echo -e $usage
    exit 5
fi
 
# =================================================
# ====================SETTINGS=====================
# =================================================
 
# Directory to store source archives
dg_downloads_dir="$HOME/downloads"
 
# Add/remove non-pip packages you [don't] want to install here.
# IMPORTANT: Only .tar.gz packages are supported.
#
# You must add a package URL and function pair to nonpip_packages array.
# If no custom install is required (i.e. normal ./config, make, make install will do) leave 
#   function call empty. For example:
#   "http://www.example.com/pack.tar.gz" ""
# If you need custom install calls, create a function that makes those calls in CUSTOM FUNCTIONS
#   area below. The function calls can include arguments.
#   For example:
#   "http://www.example.com/pack02.tar.gz" "install_pack02"
#   Or with a function call argument:
#   "http://www.example.com/pack04.tar.gz" "install_package four"
#
# MAKE SURE YOU USE QUOTES.
#
nonpip_packages=(
    "http://oligarchy.co.uk/xapian/1.0.21/xapian-core-1.0.21.tar.gz"
        ""
    "http://oligarchy.co.uk/xapian/1.0.21/xapian-bindings-1.0.21.tar.gz"
        "function_xapian_bindings"
    )
 
# Add/remove pip packages you [don't] want to install here.
# Version is included to ensure you get the same packages
#   whenever you install.
# Example:
#   "django" "1.2.1"
#
# MAKE SURE YOU USE QUOTES.
#
pip_packages=(
    "django" "1.2.1"
    "django-haystack" "1.0.1-final"
    "xapian-haystack" "1.1.3beta"
    "MySQL-python" "1.2.3c1"
    )
 
# =================================================
# =================CUSTOM FUNCTIONS================
# You should use variables such as $VIRTUAL_ENV.
# =================================================
function_xapian_bindings()
{
    ./configure --with-python XAPIAN_CONFIG="$VIRTUAL_ENV/bin/xapian-config" --prefix="$VIRTUAL_ENV"
    make
    make install
}
 
 
 
# =================================================
# ====================REAL CODE====================
# ===Don't touch unless you are sure you need to.==
# =================================================
# =================================================
 
# Create the downloads dir.
# Nothing will happen if it already exists.
mkdir -p "$dg_downloads_dir"
 
temp_file="/tmp/$(basename $0).$$.tmp"
 
# Build and install non-pip packages
build_install()
{
    for ((i=0; i<${#nonpip_packages[@]}; i+=2))
    do
        #pip_install "${pip_packages[i]}" "${pip_packages[i+1]}"
        #get package name
        package_name="${nonpip_packages[i]##*/}"
        package_dir="${package_name%.tar.gz}"
 
        echo "Installing $package_name to $VIRTUAL_ENV with..."
        cd "$dg_downloads_dir"
 
        # Check if directory exists.
        # -d test doesn't work on case insensitive FS's (e.g. OS X)
        if (cd "$package_dir" 2>/dev/null)
        then
            cd "$package_dir"
        elif [ -f "$package_name" ]
        then
            # if package is already downloaded
            tar xzf "$package_name"
            cd "$package_dir"
        else
            curl -sS "${nonpip_packages[i]}" -O > /dev/null
            tar xzf "$package_name"
            cd "$package_dir"
        fi
 
        # if odd array index has no function, do a standard make
        # Otherwise, run custom function.
        custom_function=${nonpip_packages[i+1]}
        if [[ "$custom_function" = "" ]]
        then
            set -x
            ./configure --prefix="$VIRTUAL_ENV" >>"$temp_file" 2>&1
            make >>"$temp_file" 2>&1
            make install >>"$temp_file" 2>&1
            set +x
        else
            set -x
            "$custom_function" >>"$temp_file" 2>&1
            set +x
        fi
        echo "...done."
 
    done
}
 
# Install pip packages
# param: $1 is the package name
# param: $2 is the package version
# The package version is required
#  to ensure the exact setup is installed everytime.
pip_install()
{
    if (( $# != 2 ))
    then
        echo "pip_install requires 2 arguments."
        return 25
    fi
 
    echo "Installing $1 to $VIRTUAL_ENV with..."
    set -x
    pip install "$1"=="$2" >>"$temp_file" 2>&1
    set +x
    echo "...done."
}
 
# Now install the packages if they're enabled.
build_install
 
for ((i=0; i<${#pip_packages[@]}; i+=2))
do
    pip_install "${pip_packages[i]}" "${pip_packages[i+1]}"
done
 
echo -e "\nInstall was successful. See $temp_file for details."

Now, what was I doing?

Installing Python 2.6, virtualenv and virtualenvwrapper on Dreamhost

2010-07-06T19:37:16Z

I came across this excellent and very useful installation guide on installing python 2.6, virtualenv, and virtualenwrapper.

Recovering MySQL and MySQLdb (python module) after upgrading to Snow Leopard

2010-07-06T00:14:39Z

Ahh, that’s a nice and descriptive title.

Needless to say, I had the darnest time doing this. Upgrading mysql was not too big of an issue, but getting MySQLdb to work was a whole other game.

Anyway, here are the steps I took.
The instructions for upgrading MySQL came from entropy.ch and are as follows:

Save your current database data before doing anything: cd /usr/local/mysql; sudo tar -cvf /tmp/mysql-data.tar data
Then download the MySQL installer from the MySQL website. At first I chose to download “Mac OS X ver. 10.6 (x86, 32-bit), DMG” which, I believe, caused all my problems with MySQLdb. So download Mac OS X ver. 10.6 (x86, 64-bit), DMG unless you know you have a 32 bit machine.
Once downloaded, unpack and install the MySQL package and even the prefPane to make your life easier.
To restore your old data, run this command: cd /usr/local/mysql; sudo tar -xf /tmp/mysql-data.tar
I’m not sure if this is needed, in the attempt that worked, I followed the instructions to the dot so I did reboot my computer (although the MySQL install didn’t require me to).
Start MySQL either through the System Preferences (if you installed the prefPane above) or by running: sudo /usr/local/mysql/bin/mysqld_safe

That’s all there was to installing MySQL. Note that in my system I didn’t have to change the ownership of the database data when it was “imported” into the new install. If you have to do that (i.e. can’t login with the usual username/password combo) see the instructions in the link I posted above.

Now onto MySQLdb. These instructions are taken from Geert JM Vanderkelen.

Download MySQL-python-1.2.3c1.tar.gz from the Python Package Index.
Untar it and ensure “/usr/local/mysql/bin” is in your PATH, if not see the above link on how to add it.
Run this command: ARCHFLAGS="-arch x86_64" /usr/bin/python setup.py build (Ensure you don’t have a 64 bit machine before changing it.)
And finally, install it with the following command: sudo /usr/bin/python setup.py install

Once all that is done. You can test it like this:

$ python Python 2.6.1 (r261:67515, Feb 11 2010, 00:51:29) [GCC 4.2.1 (Apple Inc. build 5646)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import MySQLdb >>>

If that gives you no errors, you’re set. Now get back to doing actual work .

Speeding to your Lord

2010-06-24T07:13:57Z

Speeding to your Lord

They speed by at 90 or more
While the ants crawl on the floor
They speed like from a roaring lion, tsunami or hell
While the ants are just trying to dwell
In an effort to achieve great ‘heights’
The birds don’t share our ‘insights’
The only heights they will gain; 6 feet under in self disdain
They live with the Law as they should
Or much deeper than that if they would
In their obedience they don’t fall behind
Oh mankind, why don’t you live with your Lord in mind?

The One who Created, Proportioned and more
This world for us to adore
Adore, adore but not just that
Lest the reason we’re created for, we forget
“I’ve not created man and jinn except to worship Me[*]”
What’s holding you back my dear Akhi[**]?

* Quran 51:56.
**Arabic word for brother.