Timeline for Why is using MySQL for a dictionary website a bad idea?
Current License: CC BY-SA 3.0
25 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 6, 2017 at 23:26 | comment | added | Joel Coehoorn | A relational database is a good idea. But MySql... not so much. It's lagged behind the other players for many years now. Take this list: CTEs, FULL JOIN, Lateral Join/Apply, and Windowing functions. They've all been part of the ansi standard since at least 2011 or longer, and are all part of Sql Server, Oracle, and Postgresql, but not MySql. MySql doesn't really qualify as a modern db engine anymore. If you need free or open source, choose Postgresql. | |
| Jun 6, 2017 at 17:33 | comment | added | Robert Harvey | MongoDB is Web Scale. | |
| Jun 6, 2017 at 16:51 | comment | added | corsiKa | I can see the angel investor now - "If only you would have used a NoSQL database instead of the most popular open source database in the world... we could have invested in you." | |
| Jun 6, 2017 at 15:49 | history | edited | enderland | CC BY-SA 3.0 | deleted 41 characters in body |
| Jun 6, 2017 at 13:50 | vote | accept | Aziz Az | ||
| Jun 6, 2017 at 13:42 | comment | added | Jared Smith | Someone is selling you something either directly or by proxy: "someone just sold me this super-duper NoSQL system as a silver bullet for all of my persistence problems, must tell everyone how great it is to avoid confronting the possibility that I got suckered by the shiny new thing!". | |
| Jun 6, 2017 at 10:13 | comment | added | Voo | @Brandon The sad thing is that the whole "NoSQL is so much faster" claims usually boil down to some theoretical explanation of why they ought to be so much better, but in practice that does not even apply for many real world scenarios. See e.g. here. Their used benchmark suite is open source and available on github too. Hell CERN manages their PB of data with an OracleDB just fine. | |
| Jun 6, 2017 at 9:07 | comment | added | mcfedr | People are telling you this because NoSQL is cool these days. You should properly evaluate the pros and cons and try to ignore the cool factor | |
| Jun 6, 2017 at 6:11 | comment | added | Kilian Foth | The idea that NoSQL stores are more efficient applies for writing, but dictionary look-up is vastly biased towards reading (English doesn't change that fast). Also, they are predominantly for extremely large data sets, but English isn't that big. You're getting bad advice. | |
| S Jun 6, 2017 at 5:20 | answer | added | joel.cass | timeline score: 2 | |
| S Jun 6, 2017 at 5:20 | history | protected | CommunityBot | ||
| Jun 6, 2017 at 5:15 | review | Close votes | |||
| Jun 6, 2017 at 15:49 | |||||
| Jun 6, 2017 at 1:53 | comment | added | Brandon | I'm so tired of people saying relational databases are bad and MongoDB is better because it is faster. That's like saying cars are bad and we should use airplanes because they travel faster. My advice is to ignore advice like this. | |
| Jun 6, 2017 at 1:48 | comment | added | Wildcard | Regarding "MongoDB is much better for performance"—as an unmodified statement with no scope clarification this is rank nonsense. For an example, see Command-line Tools can be 235x Faster than your Hadoop Cluster (which I came across from a link in The Website Obesity Crisis). | |
| Jun 6, 2017 at 1:03 | comment | added | Tim | Have you considered that some words have multiple definitions? Take for example "set" with 430 odd meanings. | |
| Jun 5, 2017 at 23:27 | history | tweeted | twitter.com/StackSoftEng/status/871871214603689984 | ||
| Jun 5, 2017 at 22:59 | comment | added | Greg Burghardt | I think we need to be careful with this question so it doesn't get closed, because software suggestions are off-topic. However given the problem I think relational versus no-sql would be an answerable question. | |
| Jun 5, 2017 at 21:35 | answer | added | Erik Eidt | timeline score: 27 | |
| Jun 5, 2017 at 20:48 | comment | added | GrandmasterB | I don't see anything here that would make me think MySQL wouldn't work fine for that. Performance on a simple lookup wouldn't be an issue, and it has full text search if you need to go that route. | |
| Jun 5, 2017 at 20:43 | comment | added | TheCatWhisperer | The people telling you things have not done much research into this. The language with the largest vocabulary, English, has less than a million distinct words. This is well within the realm of the performance capabilities of a relational DB. | |
| Jun 5, 2017 at 20:34 | answer | added | JimmyJames | timeline score: 10 | |
| Jun 5, 2017 at 20:33 | answer | added | Greg Burghardt | timeline score: 96 | |
| Jun 5, 2017 at 20:32 | comment | added | Pieter B | It's not really relational data. Just lists of words and their meaning. It's more document-like data you're talking about. | |
| Jun 5, 2017 at 20:25 | review | First posts | |||
| Jun 13, 2017 at 23:11 | |||||
| Jun 5, 2017 at 20:22 | history | asked | Aziz Az | CC BY-SA 3.0 |