Incidentally he was also doing some pretty crazy DHTML era stuff on his last home page revision, which, despite the date on the page, I believe came out circa 2001:
Hearing "DHTML" reminded me of Thomas Brattli's old site[1] and it's successor dhtmlcentral. I learned quite a bit from those sites in the late 90s early 00s.
True, true - there was no web dev/web designer, no front end/back end... webmaster had to be a *nix admin, coder, HTML, graphic designer & security guy all rolled into one.
So true. Sometimes I tell the new kids about this in terms of the differences between then and now: "No, you don't understand, I don't know all this shit because I'm awesome, I know all this because getting a computer working and online in the late 80s/early 90s required knowing just about everything. Everything. Right down to when you needed to park the fucking hard drive." It was all one big roll.
I know the long-beards would scoff at that a bit but, you know, it is what it is. Even in the early 90s you had to know almost all facets of things just to get it working.
I don't know anything about shipping, but I wonder how much of the crews' time is spent in maintenance and repairs. Are they trained to fix common issues? What happens when the robot boats break down?
Or are they just there to fill minimum crew size regulations and babysit ships that mostly drive themselves anyways?
Speaking as a former ship engineer (and designer), the engineering crew is mostly there to operate and monitor the plant. This is done from a computer. Almost everything in a modern engine space is automated or can be operated remotely from the engineering control station (the aforementioned computer). That includes opening and closing valves, starting up and shutting down systems, troubleshooting, and so on.
Unfortunately some times the sensors or actuators break, so an engineer will have to do some manual work. I suspect older engineers are more likely to do this, because they still think their own intuition is better than some algorithm (case in point: see quote in article from the union representative).
As for maintenance, it gets done in port. There are times, however, when things break horribly, and the engineers have to fix it themselves while at sea or find a work-around. With a drone fleet, one way to deal with this may be to have engineering teams on stand-by, ready to helo to a broken drone ship and do repairs.
> one way to deal with this may be to have engineering teams on stand-by, ready to helo to a broken drone ship and do repairs.
If the ships travelled in groups of 3+ then in the event of a failure the other two ships could tow the failed ship? Is this feasible? I honestly have no idea.
Not with current ship designs. It takes an enormous amount of power to tow a loaded cargo ship. They're built with some power redundancy, but not enough extra to tow another loaded ship.
Adding to gk1's comment. Adding the power redundancy needed to tow another ship would obliterate all the fuel savings these ships (potentially) can achieve.
The ships are on autopilot in deep ocean. International regulations say you need a (human) lookout, but the main job of the watch officer is to be there to handle problems and make sure that the ship is where it's supposed to be.
But most of the rest of the time is spent doing maintenance. The ocean is a very harsh environment and things are always breaking/needing attention due to constant motion and corrosive salt air.
I don't know that there would be an overall savings since if routine maintenance isn't being done underway, while the ship is making money, then it has to be done in port when it's not. So now you have a long period when your expensive asset isn't turning a profit. I haven't been in the industry in a long time, but I was taught that the second highest operating cost of a ship at sea was wages (fuel is first). As a result, ships are essentially understaffed to keep those costs down. I had to laugh at his "quality of life" comment. When has any shipowner cared about his crew's quality of life? These are people who literally fall asleep on their feet from exhaustion.
I wonder what the payback period will be. You're removing the salary cost, but replacing it with more automation that has to be extremely high reliability due to the environment and distances traveled.
Aidentified (https://www.aidentified.com) is a fast growing Series-B company working on big data and machine learning challenges.
* Front End Engineer: https://careers.aidentified.com/jobs/67199-senior-front-end-...
* Data Engineer: https://careers.aidentified.com/jobs/66399-senior-data-engin...