Live data from Hacker News

How we built Hamiltix.net for less than $1 a month on AWS

blog.badsectorlabs.com

71–77 of 77 posts

Re: How we built Hamiltix.net for less than $1 a month on AWS

#71
post #66

I can modify the price in the checkout by changing the URL parameter. Really?

Sure you can overpay if you want, and you can underpay and not get the tickets you are trying to obtain fraudulently. It's only a problem if the vendor actual sends you product that you didn't pay for. This isn't an instant digital download.

It is checked server-side. I understand why people are concerned that the price is passed in a way that they can see vs grabbed via javascript or another method like most other sites, but that doesn't change the responsibility to check the data server-side. I can edit my POST requests to Amazon but they better not sell me stuff for $0, and neither will this site.

Re: How we built Hamiltix.net for less than $1 a month on AWS

#72

Earlier quoted context omitted.

I agree, the param passing is not ideal. Session storage/local storage is another option.

Everything that can be modified by the user is not suitable. The price must be stored on the server side, everything on the client side is subject to modification.

Storing it, or passing it by the user can be suitable if the data is signed

Re: How we built Hamiltix.net for less than $1 a month on AWS

#73
> We also made the decision to not use a javascript framework for the front end, mostly because they are incredebly [sic] complex and some people suggest they are all terrible (or maybe great?).

Not to mention that for back-end developers, the learning curve is steep and ever-changing especially when one considers the "build" process. Is it grunt, or gulp these days, or webpack, or NPM, or some Frankenstein combination of all of them?

Re: How we built Hamiltix.net for less than $1 a month on AWS

#74
post #38

Earlier quoted context omitted.

I just run a server on scaleway for 3€ a month and do everything there. That gives me total control over my things.

Scaleway are horridly unreliable. I run a SaaS business, and hosted first with DigitalOcean which amounted to roughly one outage per month. I then tried hosting with Scaleway for a few months to cut costs. On Scaleway, I averaged 53 outages every month. Average response times were slower. Longer outages meant I had to manually hard-restart the server, which corrupted all the database indexes. Scaleway also has a huge…

I've got a few VPSes on Scaleway, and I've had 1 outage over the last year and a half. May just be down to luck

Re: How we built Hamiltix.net for less than $1 a month on AWS

#76

Earlier quoted context omitted.

is there something with in-depth info focused on Gun's graph database abilities?

We're in the middle of upgrading our docs/resources, maybe this is helpful? http://gun.js.org/docs (note: some pages are broken) We have pretty in depth conversations on the chat room too, here https://gitter.im/amark/gun but I'm more than willing to answer any other questions you have that might not be covered!

well basically I would just know where it falls short of Neo4J's capabilities, and if it is possible to have more than one graph running at the same time.

Re: How we built Hamiltix.net for less than $1 a month on AWS

#77

Earlier quoted context omitted.

We're in the middle of upgrading our docs/resources, maybe this is helpful? http://gun.js.org/docs (note: some pages are broken) We have pretty in depth conversations on the chat room too, here https://gitter.im/amark/gun but I'm more than willing to answer any other questions you have that might not be covered!

well basically I would just know where it falls short of Neo4J's capabilities, and if it is possible to have more than one graph running at the same time.

@bryanrasmussen , ah yes:

- You can have more than 1 graph at the same time, yes.

- Storing data on edges can be done in GUN by having an "edge node", Neo4j has a special built-in edge node which makes things easier (including bidirectional edges, which can be done in GUN, but requires making an edge node).

- Neo4j's Cypher query language has a lot of built in edge search algorithms, we have not (but plan to) built these yet.

- Neo4j has a very nice dashboard, our community has built a couple similar tools, but nothing as comprehensive.

Summary: Neo4j, being over a decade old, has a lot of features that we haven't built yet.

Why use GUN instead? The biggest kicker is it being based on a P2P/decentralized (Master-Master) architecture, which makes ops/scaling much easier and a ton of other things. If you don't need this, probably stick with Neo4j for its feature set.

Post reply on HN