Ilja Coolen \ ICSS

I travel quite a distance by train on a daily basis from my home town all the way to Utrecht and back. It’s a two hour train ride (given there are no delays) and I usually spend that time doing work, either for myself or contract work.

Most of the time there’s no problem, and delay is no more than 5 minutes. Sometimes however, when an accident happens, delays shoot up to two hours. Yesterday was such a day.

Between Best and Eindhoven there was a collision with a person, at about 16:51. Of course this will not have been a pretty sight and I do feel sorry for the victim, but an accident like this during rush hour really is bad timing. In the Netherlands, there are no real other alternatives but to try to travel through others stations. This will not save you any time though, you might just as well wait at the station you stranded for the blockage to be cleared. The NS (Dutch rail road services) usually tries to transport stranded people by bus to the next sensible train station, but it takes about an hour and a half before the first bus arrives, because they first have to call the local touring car operators, who have to assemble their troops of bus drivers. By then, the trainstation is flooded by commuters, who all want into that first bus.

I usually try to find a couple of people who want to share taxi fares. Most of the time this is way faster than trying alternate train routes or the bus. And time is money, so that taxi fare doesn’t seem to be all that expensive after all.

Since I have a monthly subscription to the rail road services, I never tried to get my train fares reimbursed, but I think I should start gathering NS delay information and claim a refund. It’s my right just as much as it is for the people who buy tickets.
You would think the introduction of the OV-Chip (public transport chip card) would make it possible to automatically get a refund if you are plagued with a delay. They can track your movement, where and when you took what bus or train (provided you didn’t get an anonymous chip card). I can imagine automatic reimbursements or refunds will never be part of the automation companies implement. It would probably cost them too much.

There was a storage administrator, whose life was perfect.

He managed a multi-vendor storage environment, where all the operating systems natively provided hardware independent multipathing to those RAID replacements the RM.com bloggers designed in the earth year 2006. All the storage subsystems available on the market were supported on all operating systems, and all migrations from one vendor to the other could be scheduled and completed without host/application disruption. He never has to convince any server operator to upgrade the multipathing software or have him remove any vendors specific multipathing software, reboot, install a new vendors specific multipathing software and reconnect the new vendors disks. He has time to spare, because he doesn’t need to plan migrations during the weekends. The operating systems he uses natively support migration of data from one disk to another (kinda what AIX already could in the early years of the 21st century). No matter what vendor. All the applications his company uses are replication aware, and have a way of recovering from a fail-over to a remote site. He has got so much time to spare, that he started teaching history lessons to high-school kids. “Good morning class, today we will discuss the archives of RM.com from planet earth, from back in the days people still used spinning hard-drives in their desktop filling computers and floor filling server racks.”

OK, maybe that still isn’t perfect. There’s no challenge in a perfect world. But I’d settle for the multipathing future.