On 10/17/2013 5:56 PM, Anderson, Kurt wrote:
Turning the table on that conversation is doable.
Make your own company understand and value the power of the IBM i. It's really an awareness thing. Our own CEOs could very well be the person sitting over cocktails talking about their own machine and system. Even if the machine isn't new, it's not like it's sitting idle with no development, right? I'm sure there are some CEOs who are keen on the latest and greatest, but most CEOs are also keen on ROI.
Sure, easier said than done, but it can be done.
I am obviously not a people-person; my ability to emotionally sway
people usually runs in the direction of annoying them, but I have had
some limited success by directly asking what exactly is it that one
wants to do with IBM i.
It turns out that so far anyway, the only thing I can't do is run .NET
and SQL Server. That can be important if Somebody in the Corner Office
has settled on a .NET package. But if that's the case, then the
decision has been made, and Someone is looking for validation rather
than a genuine analysis.
IBM i does TCP/IP. I read and write the Excel spreadsheets that the
other departments pass around as a 'database'. XML and JSON aren't that
hard. I provide stored procedures to our web people all the time so we
can easily extend our corporate reach to the web. Our version of DB2
does everything asked of it.
But there is more to application development than the platform. A
former boss was astonished to see that I use a GUI IDE to develop RPG
programs. He didn't even know that such a thing existed, much less the
fact that IBM has been selling one for at least a decade. The web group
were universally surprised that I'd even heard of TDD; when they found
out that all the stored procedures I'd developed for them had tests
written for them, they were genuinely shocked. I can read most Java and
even write a bit now and again, which helps quite a lot when considering
the cross platform ramifications of an architectural decision. I use
subprocedures inside of service programs to make rickety old programs
more stable and easier to change without breaking something else.
That's huge, because the reluctance to make changes is what will
certainly doom us as midrange programmers. 'Oh we can't change the
database because we have to identify and recompile hundreds of programs.
It would take months and then we'd need to carefully test everything.'
--Native RLA Programmer. 'I'm sure we could theoretically put that core
application on the web but our security model hasn't been looked at
since we moved off the S/36 and it'll take months to properly analyse
and modernise.' --AUT(*CHANGE) system admin. 'I'd love make an
executive dashboard but RPG doesn't have a GUI.' --Green Screen-only
The platform; hardware, operating system and even the IDE are fully
capable of creating modern applications. RPG is fully capable of
participating in the mesh that current applications use to pass data
between each other - I can get the weather information from the National
Weather Service, my package information from the USPS and UPS and Tweet
the status of them both. Using RPG. If that's not your cup of tea,
there's Java, Python and ol' reliable, C. We have Perl, PHP and soon
Ruby at our disposal. We can awk, grep and sed like everyone else. The
platform is up to snuff.
It seems to me that the larger question is: Are we up to snuff?