[AusNOG] Off Topic - Brisbane recruitment recommendations

[Andrew Khoo]

someone sage once said here that "learning how to code today is like
learning to be a welder in the 70s".

it's better to be the person that manages the coders, or provides
architectural leadership to them, than to be a coder.

and you are fully correct re being able to adapt and be "one" with the
environment.

as an example, despite all the fuss today about microservices et al, it's
nothing more than SOA on an enterprise bus rehashed.

folks have been automating stuff for a long time, bash+cron does wonders,
or even more modern rabbitMQ+sinatra etc does. migrating to some CI/CD
"pipeline manager" like Travis/Jenkins is not hard once "first principles"
are understood.

folks with smarts translate quickly to different environments.

re the comment that all "clouds" are different, i have ansible playbooks
now that take different roles to emit instances to EC2, GCE, Azure and
on-prem (KVM, ESXi). nothing too magical or difficult.

to address an issue that Mark Newton has identified, the need to "move up
the stack" for network engineering folk is more and more prevalent today.

if you are still turning up BGP peers by hand, or adding customers without
some form of automation, then i fear you are doomed.

the net result appears in the market to be less and less work available for
"junior" folk, as most of the "senior" folk have seen the light and
automated to make their lives easier (and along the way take on roles like
SREs).

if anybody genuinely want to hire folk and want to take the effort to skill
them up, folks like http://www.thenewkid.com.au/ can probably help.

however, this being Australia, i rarely find employers happy to pay $$$ or
spend time to skill up their staff.




















On Fri, May 13, 2016 at 11:37 PM, Jason Mikronis <jason at ausbbs.com.au>
wrote:

> “and with enough applied skill to attack any problem from that "first
> principles" understanding.”
>
>
>
> Hit the nail squarely on the head here.
>
>
> “Great” IT people are “one” with their machine – they know how to “speak”
> to it and obtain the results, any result, because almost all “new” concepts
> can be reduced to basics like “passing messages” or “encapsulation” etc.
>
> Jason
>
>
> *From:* AusNOG [mailto:ausnog-bounces at lists.ausnog.net] *On Behalf Of *Mark
> Newton
> *Sent:* Friday, 13 May 2016 8:14 PM
> *To:* chad at cpkws.com.au
> *Cc:* ausnog at lists.ausnog.net
> *Subject:* Re: [AusNOG] Off Topic - Brisbane recruitment recommendations
>
>
>
> On 13 May 2016, at 7:02 PM, Chad Kelly <chad at cpkws.com.au> wrote:
>
>
>
> On 5/12/2016 10:07 AM, ausnog-request at lists.ausnog.net wrote:
>
> Most people are having trouble at the moment.
>
>
>
> It is an interesting time that I've spoken about at Commsday events,
>
> talking about the void of skills in certain areas:
>
>
>
> - Cloud specialisation and experience (AWS, Azure, Google Compute)
>
> - DevOps skills - Ansible, Puppet, Chef - with deployment experience
>
> - Network (Cisco/Juniper/etc) + Linux + NetDevOps experience
>
> - Anyone with actual experience on SDN
>
> - IoT infrastructure experience - even using the AWS/Azure tools
>
> - Languages like Golang
>
> But these are all different skills.
>
>
>
> How many skills do you have? More than one?
>
>
>
> Your not going to find an individual with skills in every single platform
> and every single operating system as they are all slightly different.
>
>
>
> This is a consequence of mismanagement in the IT industry. It's a growing
> problem.
>
>
>
> It is my belief that quite a lot of the specialisation that exists now is
> a dead end. We actually need generalists in a Cloudy Cloud McCloudington
> world. People who understand concepts across a broad swathe of many IT
> disciplines, and how they interrelate to each other; and with enough
> applied skill to attack any problem from that "first principles"
> understanding.
>
>
>
> Even the newer Cloud platforms such as AWS and Google Compute vary
> slightly with what they can do and how they opperate.
>
>
>
> They all use the same concepts. A generalist can write you an API server
> front-end which can drive the provisioning pipelines on both of them via
> the same client-facing API, and write monitoring plugins to import
> performance metrics from both of them into the same Prometheus instance.
>
>
>
> A specialist will say, "I'm really good at AWS," but doom you to
> maintaining two incompatible OAM systems if you ever decide to put some of
> your workloads in GCP.
>
>
>
> You are better off building teams and having people with specific skill
> sets working on particular projects.
>
>
>
> People who believed that statement in the 1990s spent tens of thousands of
> dollars on Novell or Windows for Workgroups vendor certifications, then
> went into full revolt when Windows NT came along and their "specific skill
> sets" we're suddenly obsolete.
>
>
>
> If you want to get ahead long-term in IT, be a generalist.
>
>
>
>    - mark
>
>
>
>
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20160513/b19df4f4/attachment.html>