April 11, 2019 at 12:00 am
Comments posted to this topic are about the item Computer Disruptions
April 11, 2019 at 1:53 pm
I think yes, such disruptions will become more common in the future. In part because, airlines aren't the only industry still running on mainframes, and as those systems get older, failure becomes more likely. BUT the businesses don't see the issue, likely taking a "if it ain't broke, don't fix it and don't plan on it breaking."
Add to that, I would think the number of people who know how to both support the software on the mainframes and support the hardware are starting to dwindle.
So, at some point, a mainframe will fail (software or hardware,) and there won't be someone who knows how to fix it (or at least not readily available.)
April 11, 2019 at 1:59 pm
I do worry about that, though also the move to newer, distributed systems causing issues as we try to integrate things together. It's a tough place to be since we've tried to scale up for so long and I think now airlines are scaling out more to meet demand. Shame we haven't continued building bigger scale machines at least for some problems.
April 11, 2019 at 2:03 pm
Good morning Steve! When travel becomes a weekly issue with multiple legs on the journey, such scheduling issues can indeed become a major problem. I figured out one time that I spent more time in airports and airplanes per week than I did being at home. That's sad!!
I also share your concern with all of the shortcuts that companies take with system development and deployment. Too often it's because of a lack of proper analysis before decisions are made, deadlines made without regard to the development/testing/deployment process, and scope creep where additional "requirements" are placed while the project is in progress (of course, with the same deadlines and resources).
As with most other things, software development involves a correlation between three major things: time, money and quality. The two that typically get cut are time and money which results in a severe reduction in quality. It never ceases to amaze me how companies can justify this and yet are shocked when quality suffers.
LinkedIn: https://www.linkedin.com/in/sqlrv
Website: https://www.sqlrv.com
April 11, 2019 at 2:33 pm
No doubt widespread traffic congestion would occur on highways and inner city streets globally if a technical glitch (or hack attack) disrupted smartphone GPS capability. Imagine it's 4:00pm mid week and suddenly Google Maps, Wayz, Garmin, and Uber's app stopped working.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
April 11, 2019 at 2:51 pm
@Aaron, I think there's certainly lots of truth here, but I also know that complex systems are very complex. No one really knows all the rules, and it can be hard to ensure new changes follow those rules. I would say that I think unit testing adds a level of documentation and validation that helps to ensure regressions are limited, but in some cases, it's really hard.
Going through the migration here has been eye opening on just how complex this system is.
April 11, 2019 at 2:57 pm
Actually, issues with GPS this month: https://www.nytimes.com/2019/04/10/nyregion/nyc-gps-wireless.html
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply