3 ms·

Ask HN: Does MySql/Postgresql + Ruby/Python/Groovy solve 80% of IT problems?

• jacquesm 17y ago
  By the time you run out of steam on a fair sized server with a few slave nodes thrown in you're ticking over at 1M uniques daily, with some work you can stretch that quite a bit further (sharding). Of course this depends on how much of a load your average user generates. The database is a bottleneck, sure. But usually you hit other roadblocks on the way up that are a lot harder to solve than how to make your database scale. To put it in very pragmatic terms, on the web there are currently 1,000 companies with more than 1M uniques daily, if you're going to be one of the lucky 1,000 then you will haver a scaling problem some day. But that scaling problem will be a relatively minor problem, plenty of people have solved that and documented it. Going noSql is one of the ways of getting around that but there are several ways of achieving that.
• chipsy 17y ago
  No, because these things are not "solutions." They are technologies. You apply the right technology to the right situation. It is true that there are technology fads, but the fad technologies are often linked to growing awareness within the industry about the concepts they employ. For example: C++ -> Class-based object orientation Java -> Virtual machines Rails -> Web frameworks NoSQL -> Scalability strategies Erlang, Haskell -> Concurrency strategies So really, the appropriate way to ask this question is, "are these concepts enough to solve 80% of IT problems?" Once the concepts are available to the mainstream, switching the specific technology is not a huge problem. It's true that in the process of taking on fads, people tend to lose their way and become zealots, getting stuck on one concept as the end-all be-all, but that's true of most human topics. I would venture to guess that at this point in time we're actually seeing a whole explosion of concepts thrust into the mainstream from many directions - some aged and well-known to academia, and others the result of front-line efforts.
• bayareaguy 17y ago
  It depends what you mean by the terms "solve" and "IT problems". The things you mention address at best 5% of some of the IT research problems of a decade ago[1] and even if you mean IT problems in the more mundane sense then the answer is still no since a) most of those are really organizational problems and b) new IT problems will appear as soon as your older ones are resolved. However if the question you're really asking is "can those technologies help me reduce my effort or cost by 80% ?" then thanks to advances in hardware, services and open source in many cases the answer is yes. 1- http://research.microsoft.com/en-us/um/people/gray/talks/gray_turing_fcrc.pdf http://research.microsoft.com/en-us/um/people/gray/talks/gra...