Yeah, works better... after repeating myself five times now, we're still not on the same page, but this is working better?
No, I understand what you are saying. In practical use cases I don't believe it. I don't believe that my method is wrong and that yours is better. This was a SQL schema question, so why does it not make sense to address the schema issue?
Entity-relationship modeling does address the schema issue; The whole point of doing it is to create a good schema! The ERD would have shown that logically we were on the same page, but in terms of the physical implementation, you want to create artificial keys. I had to explain THAT several times. Moving on...
Take the specific case and then drill down into what is going on. Theory is all well and good but what happens in practice is far more important. The world doesn't run on theory.
Yeah, just earlier today I was thinking that...
Google... probably just guessed a good search algorithm
IBM... O(n^3) sorting would have been good enough to get them off the ground (ie they didn't really
need a fancy sorting algorithm, and besides, they wouldn't have even KNOWN that it was a FANCY algo either! Boy ignorance really is bliss isn't it!)
Akamai... The whole thing with distributed hash tables was probably just a lucky bug in their code
iRobot... Nope, no theory stuff there; That shit is as "random" as it gets, right?! (and elephants not playing chess... pfft!)
Intel... They were playing with transistors and just happened to make an ALU; The whole boolean logic thing came decades later to punish the rest of us for being late to the party.
Isn't the term APPLIED SCIENCE confusing? To what are these scientists applying? Did you know that over 33% of US GDP is attributable to quantum mechanics? Seems strange that SciAm was able to relate a theory to the real-world in such a manner when the world obviously doesn't "run on theory". Seriously, you need to sit down and think about what you just said, and for your sake, I hope you're more considerate to the theories of say physics than you are towards math and computation. You may just find out the hard-way that Newton et al "theories" about gravity, electro-magnetic force, and energy actually have some relevance to the "real world".
I am going to ignore the snipes at my typing. I'll try to use a NFA or something for you next time.
Honestly, I couldn't figure out what you were trying to say about search...
Who said anything about Flickr data? Do you work there or something?
No I don't... but the first line of the first post in this thread:
I am in the process of creating a custom photo gallery system for a website, and need a little advice on how to set up the tables in mysql.
Trying to keep it relevant to the topic in question... the VOLUME of data has a large impact on how the queries perform. Flickr is basically the same thing the OP wants(ed) to create.
I find it a bit ironic that someone who was just arguing "real world" vs theory would suggest (1) testing with a volume of data that has to be at least 100K times the volume the OP would ever be able to generate or (2) conclude that Flickr is what the OP actually wanted to create. Being incredibly prone to realism at the moment, I would even suggest that a site like Flickr probably use something a bit less naive than just a simple relational database and sql query... just a theory.
As all of the data is being generated, we can test with different volumes of data. We can pick numbers of records that we find to be meaningful, perhaps, 1K 100K 10M and 1B?