Show HN: Review Signal - Real web hosting reviews based on social media comments | Hacker News Reader

Show HN: Review Signal - Real web hosting reviews based on social media comments

reviewsignal.com

19 pointsohashi15 comments

It seems like a good idea, but I suspect hosting companies could either setup fake twitter accounts or get their employees to spam positive reviews. Where as normally they have the dirty practice of logging into a review site and generating plant reviews (something that shows intent), this way they can just claim "Hey, we were just posting positive reviews to all 10,0000 of our own twitter accounts, we had no idea they were being used on Review Signal!".

They say "All messages from people financially or otherwise connected to the company will not be counted in the reviews.", but how do they really know?

Honestly, like SEO, it will be a game of cat and mouse. It's pretty easy to see big changes if a company gets a lot of tweets. Then it becomes an investigative matter and figuring out how to find real vs fake. It's not perfect, it never will be. But I think it's a step in the right direction (I could be biased ;) )

This fails for me.

I selected Developer, budget between 101-250, for "something else", while being comfortable managing it myself.

That something else is all my projects/websites essentially, which all differ a lot.

I doubt running that on Shared Hosting is a good idea?!

Besides that, awesome job - I like the idea, most review sites are filled with fake reviews just to make the website owner make the most via the programs that pay most..

You're absolutely right. Honestly, the hosting recommendation tool is designed more for people that don't know anything about hosting right now. Making good suggestions for developers is hard. It's one of the things that definitely needs improvement.

That said, given what you put in, what sort of recommendation would you expect/want? I have some ideas but I would rather hear yours first before injecting my own opinion.

Well, selecting other - maybe a longer dropdown - entering how many sites (and what kind), bandwidth usage, visitors - rough figures obviously.

Some awesome features would be entering X amount of ip's (if you want unique or multiple), and maybe also add something regarding main demographic (should I go for a US provider, Asian, or European).

Should I go for a VPS? Or a Dedicated server? I know, it's tricky to calculate all of this - but it'd be awesome!

Last but not least, I'd only show these options if someone were to select a checkbox for advanced options.

Thank you for the ideas. I have added them to the feature list. I will see what we can come up with based on your suggestions. The difference between vps/dedicated/cloud is really hard to recommend these days. It's also getting all mixed up, making recommendations even harder. Any ideas on that?

This seems handy. I already knew a lot of the companies within the bounds of the configuration specs I listed, but found a few new ones.

I'd love to see digitalocean.com on there (maybe I missed them?). No affiliation ... just saw someone link to them recently and would be curious how they stack up with the others.

Seems like the top listing for most categories is mostly there because they seem to do informal support via twitter leading to people responding happily. I'm not knocking them, I've never heard of them and they certainly could be great, but I'm not primarily looking for a hoster that uses twitter as a ticket system.

That's a fair criticism and concern. From what I observed most companies are interacting on Twitter and some do it better than others. Do you have any ideas on how to improve the system? I know my way isn't perfect and I am always looking for ways to make it more accurate and representative.

I'd be curious to see what sentiment analysis of web hosting talk (or similar?) yielded.

finally a site with enough data to make a credible recommendation on hosting companies, any chance you'll cover overseas markets at some point?

Right now it's very US centric. As far as covering other countries, possibly on other english speaking countries like the UK. I am not sure I am ready to expand into analyzing other languages, so that would be a major issue for expanding to many markets abroad.

how does it classify tweets into negative/positive?

will this last long? with twitter's new "rules"?

The change (as far as my understanding goes) is it requires authenticated requests. I don't think it should be an issue, it just means doing everything with Oauth.

Reply on news.ycombinator.com