WebGen Services

Yes the struggle is real!

Luckily not all our clients!

What does AI and progress mean to you?

Just a few steps - WebGen And yet the steps got bigger very quickly. I feel privileged that I grew up on a world of black and white TV’s which soon became colour. Soon they were followed by home computers what a whopping 48k of RAM and a dizzying array of 16 colours! For comparison my next computer will have at […]

We've added a What'sApp button so that you can reach out with your web development needs to WebGen{eration} - fire at will!

Passwords can be difficult to make complex but still easy to remember. One way is to write a cipher - a fancy term for basic encryption. Pick your own questions for example: Favourite colour/pet name/birth town/mother's name.

blue
diggy
adelaide
wendy

Then swap out the "easy" numbers and other characters making vowels capitals:

b|U3
d!ggy
@!d3
w3ndy

Now read vertically:

bd@w|!d3Ugdn (First 3 columns).

The benefits of this is that you only have to remember the questions and their order. (and your swapping rules) but also it's a very complex password. After a could of goes at it you will remember it anyway.

For added protection you could add the web site:

bd@w|!d3Ugdn-li (LinkedIn)
bd@w|!d3Ugdn-fb (Facebook)

What ever your method make sure that you update it from time to time. With the method above you could just choose the columns in different orders.

Stay safe!

Is it plugged in? Users hate that question but there is a reason we ask. It's called the OSI model. It isn't literal but it is a valid model. To paraphrase it is a stack that starts at the physical layer (is it plugged in!) and ends at the application layer (is what you trying to use actually working).

We had a client reach out for assistance yesterday because their sites were slow or unresponsive. Our approach was asymmetric to the model (now you know why geeks have two monitors). Basically it's top down and bottom up to the OSI model. Firstly we check things like DNS to make sure their sites can be found. All good there. The sites sometimes worked as they should and seconds later would hang and not load.

We spun up an instance in their ISP and tried to replicate their web server's experience. We had the same issues. It would work now and then and drop out seconds later. Running a traceroute to the server was flawless regardless of the web server being available or not.

This is the difference that expertise and discipline provides. Traceroute uses ICMP as a protocol and not HTTP/HTTPS. Knowing this we could identify that the problem was upstream and not with our client's servers.

As we don't have access to our client's ISP but given the on again/off again results it would one of their round robin load balancers for HTTP/HTTPS was not working (this is like when you go to a hotel and your not sure which one of the 3 staff you should stand in front of to check in).

Whilst we may seem to ask dumb questions we acytally don't. We follow strict protocols so you don't have to know about them. Sure it may be that we have a rigid language but that is so we are consistent. By all means call your mouse a hampster - seriously we can adapt to that.