640K is enough for anyone
My first “real” computer was a good ol’ Commodore 64. 64k of memory, external tape drive that ran off of “standard” cassette tapes, or the external floppy drive which was capable of 320k total storage! (Using both sides of the disk BTW.)
My second computer was bought from a furniture store in the early 90’s. Around here computer stores were still few and far between. It was a 386SX-25 with a whopping 4 megs of Ram, and a 80MB hard drive. Cost something like $2500. But man did I enjoy that machine. I had no idea what I was ever going to do with 100MB of space. DOS, Windows 3.1, and about a dozen games I had barely used half of the hard drive.
My next computer I bought the summer between my Junior and Senior year in high school. This was back in the day when I made $300+ a month and had no bills what so ever. Had about $1500 saved up in my bank account. I found a guy who was hard up for cash to pay for school, and bought a 486SX-33 with 6 Megs of Ram, and a 110MB hard drive off of him. The interesting thing about this machine was that it had four 512k memory chips in it. Those were incredibly rare, but existed.
This machine served me through college quite well. It got upgraded with a sound card and CDrom Drive. Then later on got upgraded with a 28.8k modem. THAT was awesome. I had the first one at the school – possibly the first one available to the public in the state. I was smokin’ with that computer. At this point in time I was filling the hard drive – tons of games mostly. But some graphics, midi files, misc text files and such.
A year later I got a job in the computer store. My knowledge was apparently far “beyond,” many others. So I jumped in, saved up $$$ and went to town upgrading my machine. New hard drive. Then a new mother board – finally upgrading to a Pentium 75. 16 Megs of Memory, 512MB hard drive, more, more, and more!
My computer was in the shop nearly every month getting a new upgrade. We would take the old parts and recycle them at slightly reduced prices into the used pile and people would instantly buy them. Was good in those days.
Nowdays I’ve got 18GB in each of my two “big” servers, and I don’t really consider that being very much. I have three 60 GB hard drives sitting around waiting for me to install into one or both of those servers, I have 40GB on my Powerbook. Then there is all the other machines I have, multiple gigs per a machine, multiple machines.
The funny thing is – that in this day and age, the bulk of the hard drives space I use it dedicated to software. I have fairly low-key needs personally, an OS, some word processing, my financial software, a web browser, maybe a game or two. I’ve got about 11GB of MP3’s, which represents about 1/3 of my CD’s after I ripped them and cleaned out the duplicates/bad stuff. Some pictures, mostly backgrounds and such (OSX’s automatic background switcher is pretty cool.) And then I’ve got about 4 gigs of various Ebooks – I’m such a pac-rat when it comes to knowledge.
What brought this whole thought up was a piece on NPR this morning talking about how search engines work. They talked about Yahoo going through and having humans validate web pages, then storing them into a database to be looked at. I started thinking about the numbers of bookmarks I have on my computers – a couple of thousand probally. So for them to have the books marks, AND be able to cross reference data in those pages, AND be able to access it millions of times per a second is simply astounding in terms of data storage.
At work our data needs range into something like a mere 8 terabytes. Most of it graphics and designs. Yet every bit of it is backed up nightly and sent to off-site storage. Quite another feat in itself. (I don’t work in that group – but I assume it’s incremental backups of changes only to keep the actual amount needed down much further.)
This all becomes quite astounding when you consider that we’re probally pretty typical of a large company. So thousands of companies around the world are archiving data. Governments are archiving data, and each person who owns a computer is archiving data seperately. The amount of data on this planet must truely be astounding. I imagine that The Ancient Library of Alexandria wouldn’t even make a dent on it.
So the real question is – does that much data REALLY need to be archived? If we remove duplicated data, how much of it is actually, truelly needed? I’ve been thinking of coming up with a data model that allows anyone in the world to have access to paticular data that is needed, shared out essentially much like a Peer to Peer network. Add in permissions and ownership rules for sensitive data and it could remove a companies large cost of data backup, while allowing knowledge to be shared easily.
As a home user I could have a group of bookmarks based on data type to certain files that contain the information I need/want. But it would be physically located in a central repository, much like a modern Library of Alexandria. If it’s a copy righted work, you could pay a small one time royalty fee to the orginial owners in such cases as novels or music. You get your limited rights to the IP which pretty much means you can read/play/use it as you see fit minus the normal rules for playing in public, etc. You could even transfer or sell those rights to somebody else if you wanted, and if the orginial copy right allowed it.
So it would break down like this:
Owner of data defines who can look at it. Sells licences if needed, and can decide if the licence can be transfered by the 2nd party.
Buisness could look at data and share among them selves, or other partner companies. They could sell licences to that data to interested 3rd parties who might add to the data.
Seperate flags could be used to say X data links to y and z data and that linked has been explored x+1 number of times much like Google does now. This way you get the useful data you want, plus other links to it that maybe related.
This would cost a of money to setup, but wouldn’t it be the way to go in the long run?