[thelist] What's a Programmer To Do?

Kevin Timmins kipper_timmins at live.co.uk
Mon Apr 6 12:53:13 CDT 2009


Hi Jack,
I agree with you here, any single coder should really have their own 
personal arsenal of code/framework/structures that they themselves have 
developed.
how many people "trust" micro$oft? people i speak to fully accept that 
micro$oft build in holes to be corrupted. is this true? who knows, but if 
you have your own code then you have your own peace of mind that you 
yourself fully understand what your code is doing, why it is doing it, and 
that you have taken as many precautions as possible to limit security 
issues.

nowadays i think there are so many people interested in the idea of 
automation, but is this really a safe route to go down? if everything 
becomes automated then who fixes the automation if all anyone has experience 
with is just running the automation?

generally i hear of the likes of drupal and whilst i accept that it is a 
fair feet of coding it is essentially just building another programming 
language from a previous coding language. java is built on c or c++ can't 
remember which, drupal is built on php which is built on another language, 
all in the name of automation. and there comes a point where you simply 
loose capability to do certain things without stepping back a language and 
making a custom module or custom class etc.

certainly it is a war on preferences, but in a world without plumbers, who 
fixes a broken pipe? and lets face it, some of us here are hopeful plumbers 
of programming.
so this is just my 2 cents to the thread so far :D
cheers
kev

--------------------------------------------------
From: "Jack Timmons" <jorachim at gmail.com>
Sent: Monday, April 06, 2009 3:21 AM
To: <thelist at lists.evolt.org>
Subject: Re: [thelist] What's a Programmer To Do?

> On Sun, Apr 5, 2009 at 7:19 PM, Martin Burns <martin at easyweb.co.uk> wrote:
>
>> Which is quicker: write a (eg) login system from scratch, or take one
>> off the shelf?
>
>
> Your example isn't a good one, since a login system is pretty easy to do
> anyway ;)
>
> But, my stance is more on developing your own framework and tools.
>
>
>> Which is more likely to be mature: the system that you've knocked up
>> in a few days, or the one that's had thousands of days, and hundreds
>> of implementations behind it?
>>
>> Where's your time better spent: understanding the client and doing
>> customisations to a system that's 80+% there already, or building bog-
>> standard functionality from scratch?
>>
>
> Using a system thousands of people have access to, so not only are they
> (hopefully) fixing vulnerabilities, but (more than likely) looking to find
> methods of exploiting them. Giving the customer a bloated piece of 
> software
> that they're only really using a small percentage of?
> Just playing devils advocate, as you did previously. Really, this whole
> debate is nothing more than a war on preference. Ultimately, the coder's
> best bet lies in going with what they're best at, and trying to 
> continually
> improve themselves.
>
> -- 
> -Jack Timmons
> http://www.trotlc.com
> Twitter: @codeacula
> -- 
>
> * * Please support the community that supports you.  * *
> http://evolt.org/help_support_evolt/
>
> For unsubscribe and other options, including the Tip Harvester
> and archives of thelist go to: http://lists.evolt.org
> Workers of the Web, evolt !
> 



More information about the thelist mailing list