Friday, November 23, 2012

How internet changed the world, as seen from below

I have just read the news that a 61-years old computer has been successfully rebooted. This reminded me how lucky I was to be a direct witness of the computer science revolution that started at the end of the XX century. I am not so old to have the same age of the above cited computer, but I was born a numbers of years ago sufficient to have used punched cards (see the photo on the left). I do not want to write an essay with the detached eye of the history scholar. I just want to remind some personal recollections on how computers changed my way to work and live during just a couple of tens of years. I guess that many scientists of my age could share similar experiences.

My first experience with computers occurred in 1981, when my father bought an Apple II+ for his office. He was the first in our small city to adopt a computer for the office work and had to face with the usual colleagues who did not understand the innovation power of a computer. His colleagues told him he was a megalomaniac, that computers would not have lasted for long time, and similar amenities. Nevertheless, my father believed he was on the right track and indeed he was correct. 

I was immediately fascinated by that machine and after dinner, when my father's office was closed, I slunk in the computer room and started playing (and learning). The Apple II+ had 48 kb of RAM memory (later expanded to 64 kb), 5".25 floppy disks formatted at 140 kb, a green-phosphors 40-columns small monitor (able to display only capital letters!), and a 9-dot printer. Anyway, I learned by myself, just provando e riprovando (try and try again), the fundamentals of computer science: BASIC and Pascal languages, the disk operating system, and the logic. 

A few years later, when I started to study at the University of Bologna, I made experience with the Digital VAX 11/780 and the punched cards. The 1986 was the last year that the University adopted this computer for exercises of the students of the course of "Numerical Calculus and Computer Science". We had to accurately design our software to avoid waste of paper and computer time. Indeed, we had to prepare our set of punched cards and gave them to the computer operator, which in turn took care of running the program. There was no video output and the results were printed on sheets of paper. Therefore, it was extremely recommended that the software run correctly already the first time. Errors would have caused delays of days, because there was the queue of students. Since there was no video, we had to use some trick to print a graph. The most commonly used was to print some special character (e.g. an asterisk *) with a number of spaces from the left border of the printer paper. The number of spaces was equal to the Y coordinate of the function to be printed or any properly scaled value. The X axis was the horizontal line of the printer paper, with one line per point. An example is displayed in the image below.


My first really personal computer, i.e. bought with my own money, was a PC-compatible with an Intel 8088, 640 kb RAM, 2 drives for 5".25 floppy disks, and a 20 Mb hard disk. I bought it in 1988 and it was as large as a travel case. I used it to print my first doctoral laurea thesis in 1990. At that time, there were not yet many special characters (e.g. greek letters), so I left empty spaces in the printed pages, which were later filled with dry-transfer letters. Windows was still at an embryo stage and the Apple Macintosh was too expensive for a graduating student. The software code I developed for the thesis work was written in FORTRAN77 and run on a MicroVAX II of the University, with VAX/VMS operating system, an unsuccessful alternative to UNIX. I logged in through a "dumb" terminal Digital VT100. I used the same computer for the Ph. D. work, but at least the PCs and software were developing quite quickly: 80286, 80386, Windows 3.0, and software for terminal emulation. When I wrote the Ph. D. thesis in 1993, I already used Microsoft Word 2.0 for Windows, equipped with the Equation Editor that allowed to write equations and special characters in a nice and readable form. 

In 1993, I also started the collaboration on planetary science with the Meteors group at the Institute of Atmospheric Science and Climate (ISAC) and I began using the internet, the web, the UNIX operating system, and LaTeX formatting system to write articles. At that time, the software for emails was just mail or Mail on UNIX, which in turn send just pure text emails. Later, I started using pine. The web browser was the NCSA Mosaic and I remind that in the very early days I dreamed to visit all the web pages in the internet. However, I immediately lost any hope as I realized that the number of web sites was exponentially growing each day (or even each hour!). 

In 1997, I started to collaborate also with the Atmospheric Dynamics group of ISAC, where I worked on numerical mesoscale meteorological models. We used the supercomputer CRAY Y-MP, hosted by CINECA, at Casalecchio di Reno, near Bologna. One run generally took time of the order of one day and one challenge for weather forecasting at that epoch was indeed to reduce the simulation time to values smaller than one day. At the end of nineties, CINECA dismissed the CRAY, and replaced it with an ORIGIN 2000, with 16 processors. However, the power of workstations was already comparable with these supercomputers. I remind that the Atmospheric Dynamics group bought a Digital Alpha workstation and we did some comparative tests with the ORIGIN 2000. The results indicated that there was no significative difference, unless to adopt a very massive parallelization of the codes.

The computer power continued to increase during the years 2000s, although not at the level of eighties or nineties. The availability of Linux and other forms of open-source UNIX operating systems, greatly enhanced the computing power of personal computers. When I contributed to the development of the data analysis software of the IBIS imager onboard the ESA satellite INTEGRAL (this time I adopted C language), the whole package and the data to be analysed could run on Linux-equipped PCs with moderate resources. Now, I use an Apple MacBookPro, which contains almost all my office stuff (data analysis software, data, latex, and other office software) and I have the advantage to bring it always with me. The advent of the iPhone has also changed the way to work and we are today living almost always connected. 

It is somehow amazing to see how life changed in just a few tens of years: in 1986, I used punched cards and there were still computers filling one room. Today, I have a much more powerful computer in the palm of my hand and I can connect to all over the world in just a few seconds. The world changed so much in so short time. Best? Worse? I do not know. Although I work intensively with computers and smartphones, and enjoyed the improvements developed in these years, I still like to work with pens and sheets of paper. When I have to carefully read an article, I have to print it. I am not able to read an e-book and love the feeling to keep in hands a "solid" book, to underline with a pencil the most important phrases, to write notes on the borders of the pages, to go back and forth along the pages. Perhaps, despite having lived the explosion of the internet revolution, I am still of an older generation, perhaps not able to fully exploit the power of computer science. Or, the world of the printed paper has still something useful and interesting to say...

Note added on 14 December 2012: here you can find a nice program to have an idea of how punched cards and computers worked during sixties and seventies. 

No comments:

Post a Comment