Previous: Chapter 5. The actors, Part II
Chapter 6. Bletchley Park
In which we get close to the heart of the mystery
It was well into April. The weather had turned lovely, the air started to have that earthy spring smell and new green appeared everywhere, when klimagalka finally discovered what had become of the materials.
@klimagalka:
@kagetsuko My friend’s colleague says the material was moved to the Computer Museum in Ekaterinburg. I have contacted them, will keep you posted. @silverstacks @lores
And a few days later:
@klimagalka:
@kagetsuko You’re not going to believe this, but they sold the material to National Museum of Computers in the UK. I’m afraid I can’t help you any further. @silverstacks @lores
@lores:
@kagetsuko @klimagalka That’s in Bletchley Park, it’s not very far from where I live. I’ll be very glad to go and have a look at it if you can make the introductions.
kagetsuko contacted the museum and they confirmed they had some materials in the archive that originally came from the Siberian Computer Center and were dated between 1970 and 1980. They agreed that I could come and study them.
And so, on the morning of what promised to be a bright day in the early May of 2027, with every bird in the dawn chorus competing for its space in the audible spectrum, I set out for Bletchley Park. Which was about 200 km as the osprey flies, but as the railway network in England is so London-centric, it was a three-hour hour journey.
The archivist showed me the materials, a few storage boxes with paper documents, 12-inch floppy disks, various kinds of tape and, lo and behold! several thick stacks of punched cards. They were well indexed and I had no trouble with the Russian captions. I found the bricks of cards used for the 1975 runs, scanned them with my phone, and returned home. I had a decoder for the BESM-6 punched card format on my laptop, so I spent the return journey decoding the scans into FORTRAN source code. Success at last! It was night when I returned, so all I did was post a quick message and go to bed.
@lores:
@kagetsuko @klimagalka @silverstacks I’ve got the source files! Will check them tomorrow.
Next morning after breakfast, I started checking the source code. It was not very large, so it was easy to work out the structure, but it was rather convoluted. Not that it was poorly written, but such coupled models have a complex data flow and very large numbers of parameters. It was easy to separate the NOAA cloud feedback code from the Soviet code, purely on coding style. However, on closer inspection, the code turned out to be incomplete. The crucial kernel was missing.
@lores:
@kagetsuko I have the code for the whole simulator, except for that crucial convection kernel. I will go back and see how I could have missed it. @klimagalka @silverstacks
Next: Chapter 7. VEB Kombinat Robotron