Hi All, I would like to ask for advice and... debriefing about choosing which tool to adopt for setting up a neo4j wrapper :)
I'd love to build an MVP as such: you have a web client, you typed in your query, and visualize recommendations against keywords, based on Neo4j. Servers are not my expertise, but I working with a prototyping matrixes and getting in love with traversing them with neo :) I wanna be able to set up my MVP and ... pitch :) Now, at server side level, even if I am not the guy who will take care of optimisation and so forth, I d like to set up actually more than a demo, that means sometime I actually picture to work as a alpha product, something to start with which can be than improved. I would like to use a wrapper for my neo4j, so that I can sent minimized JSON over the network, and not the large files which neo produces. Also, I am thinking to solutions that would allow to scale. So far, working at local level, no cloud, all on my laptop: it will likely come the time to bring the whole thing on cloud and use cloud services, but so far, please help me in choosing. My skills in coding start to be decent in python for analyzing and crunching datasets, and I got the hand dirty in Javascript libraries to visualize data, learning basics in bootstrap and jquery. I learnt a bit flask, with tutorials, and where able to expose API in a two-three days. However, it looks to me that for production you either go for Django, which looks to me much more complicated, and that setting up an API REST with flask should be more robust to scale. I got curious about node.js, because it looks like it can 'natively' scale well, and it is gathering quite a momentum for backing web app. There are also rocketing libraries out-there build in JS, my favorite choice is actually Vivagraph, now the author is also working for porting the library on npm. I understood npm modules are like extensions for node server: not clear to me how they relate with front-end visualization. However, I start to salivate the idea that, even if node.js is server-side, maybe working with libraries made in JS to visualize data, could in some how "integrate" more easily, at least as learning curve. I've been also inspired by: http://www.ebaytechblog.com/2013/05/17/how-we-built-ebays-first-node-js-application/#.VF_mzPTF9RE My question is: which tool would you advise to start work with, and which learning curve should I expect? Which main differences between neo4j + node.js and neo4j + Django/flask/python would you spot ? Which difficulties should I expect? Which framework is likely to become more popular ( also thinking for getting support or advice or tutoring)? Which books would you suggest? Also, full text indexing with regex is not super well optimized yet, and would like queries to be done on that. Is thing I should consider later on, or does this suggest a tool instead of another? Thank you for helping me clarifying some concepts! -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
