On a chilly evening in October 1995, I dialed in to my local ISP (Internet Service Provider), heard the screeching of the modem, and then launched an FTP client to upload the HTML, JPG, and GIF files from my Macintosh IIsi to create “Frank’s World.”

No alt text provided for this image

Before the Launch

It had been a long time coming, in 1994, I had heard of the world wide web and used it quite a bit using something called the “Lynx” browser — a text only browser that worked on VT-102 terminals. I had thought about creating a website, but wasn’t sure what I would build.

To anyone under the age of 30, it’s hard to imagine that there was no Google, no YouTube, no Reddit, etc. The main way to learn technical topics was to buy books. At the time, there were few titles about writing HTML. So, when Laura Lemay’s “Teach Yourself HTML in 24 Hours” came out in the Summer of 1995, I picked it up immediately.

HTML was a subset of SGML and, as a recent graduate with a computer science degree, it was not hard to learn this “language.” (JavaScript was not yet a glimmer in Brendan Eich’s eye.) At the time, there were not any good HTML editor. I remember installing and subsequently being extremely frustrated by the HTML export plugin of Word for Windows 6. HoTMetaL and HotDog were even worse (and they charged money for it.)

The idea for Frank’s World originally was a parody news site along the lines of the Weekly World News. This was before the Onion and way before Babylon Bee. The Weekly World News didn’t even have a website yet. Yes, Virginia, the web was that new!

The Launch

I uploaded the files to the 1 MB of disk space that my dial up internet provider included for free. The domain name would come a year later.

As for launching, I told a few friends and family. Most of my family members didn’t quite understand what it was. I remember being excited when Yahoo! added me to their index.  At the time, Yahoo! reviewed and added each web site individually.

More to come

I’ll be launching a limited run podcast series on the history of the site, which reflects the history of the web.

As Apache Spark is 10 years old. This article in Analytics India Magazine explores what led to Spark’s widespread adoption and what will keep it going into the future.

Dubbed as the official “in-memory replacement for MapReduce”, the disk-based computational engine is at the heart of early Hadoop clusters. Why Spark took off was because it reflects the changing processing paradigm to a more memory intensive pipeline, so if your cluster has a decent memory and an API simpler than MapReduce, processing in Spark will be faster. The reason why Spark is faster is because most of the operations (including reads) decrease in processing time roughly linearly with the number of machines since it’s all distributed.