/vr/ recommended/best of site

Alright lets reddit/twitter this shit up.

No, recommendation sites are stupid. Just look up Mobygames for tagged games.

>>3224589
Be prepared to get spammed endlessly by assholes, flooded with different spellings for each game, and redundant-but-differently-named genre categorizations (eg. "beatemup" vs "belt scroller"). Leaving it open for just anybody to submit entries is going to be hell to moderate and sanitize.

I'm down, give me somewhere to post and I'll do it, I love looking for great obscure games. Has anyone here checked out the GameBoy Wizardry titles?

>>3224601
maybe then only a relevant /vr/ thread for new additions/subtraction arguments?

i'm also fine with "this is a gay idea don't waste your time"

>>3224604
I think you're better off compiling your own list first and using threads for people to field suggestions. Keeping direct control away from us is the only way to keep it coherent.

>>3224626
well i guess this more of a soft announcement of possibly doing just that so it doesn't seem like gay ass spam or some shit later

>>3224602
I have, but not much.

>>3224589
I like the idea but if you allow GBA or PS2 it's garbage

>>3224795
i won't be allowing them

>>3224589
Topic aside, that map looks really useless. What is the point of making a map for a pretty simple adventure game, and then making it borderline unreadable by not caring to arrange it in a readable order?

Also expect to get spammed with memes like "Doom and Gimmick are best games, Castlevania IV and Symphony of The Night is the worst".

>>3224589
why prefix all your tagging with #? What about general meta data, instead of relying on tag gibberish exclusively? How are redundant/duplicate tags handled? manual fixing? aliases?
Why an alphabetic list of tags, when the tags clearly have categories? Even your Sonic example had the platform and the genre categories for the tags

>>3226298
>>3226351
>didn't read the thread

>>3226580
I'm >>3226351. Which of my comments were addressed already?

>>3224589
I had the same idea last year but I deemed it unfeasible.

Anything community driven is too hard to sustain, you spend more time editing and rejecting content than actually building new content.

You're also going to end up with a -lot- of Homogenization because most people are going to do the same things.

I wanted to have it all arranged by Platform-Genre, but also have general stuff like

>Recommended TG-16 games
>Recommended SNES games
>Recommended Platformers on the Genesis
>Recommended Party Games

My criteria was that I beat a game, or spent at least 5-15 hours if I couldn't beat it before I could give a recommendation or not. You can\'t expect the community to do such a thing. You'll end up with people making recommendation lists going "I loved the shit out of OoT but I never beat it tho" and their opinion should be discarded.

Besides, if someone goes "recommend me some SNES games", you already know what's on there.

>1.Super Mario World
>2.LoZ: A link to the past
>3.Donkey Kong Country 2
>4.Earthbound
>5.Starfox
>6.Mega Man X
>7.Super Metroid
>8.Chrono Trigger
>9.Super Mario RPG
>10. Secret of Mana

It's like clockwork. You can go ahead and do it, but to set yourself apart you'd need it to either be different (in which case you run the risk of being contrarian and have people dismissing your site), or bring up "le hidden gems" like MetalJesusRocks does (and fails at).

I wanted to make it, but the time investment required to actually get enough content for it to be sustainable is disgusting. I was going to have mini-reviews and why it was worth playing for every list.

It's probably easier to just learn how to make youtube videos.

>>3226601
if you're brainstorming game ratings and recommendation lists, here's a suggestion: let users, signed up or otherwise get a selection of pairs of video games, and for each select which they liked more. The pairs are random. Different platforms, genres, years, just fully random. It's kind of hard for people to compile their top-10, and you end up with generic lists, but it's super easy and fast to select which of two they liked more. Apply an ELO-like rating, which does not rely on the number of ratings, and you get good results quickly.

>>3226607
and of course, with each game having its own rating after all this is over and done, you can trivially group them by genre, platform, year, whatever people are looking for. The rating was entirely neutral. Only problem is, you will need a lot of ratings. Could do that with signed up users and achievement/seniority whoring, if you're serious about it

>>3226613
>>3226607

The idea was to have personally curated lists, so if you found a contributor you jived with you could "follow" them and their content as opposed to other people who you didn't like as much.

So that makes it inherently hard to consistently get good content, especially if you aren't really paying people.

The biggest thing I felt was holding the site back is it's inherently clickbaity. No one gives a shit about Top 10 X's or wahtever in any real capacity, and you're building an entire site off of it.

Sure, HG101 is heavily flawed, but at least the articles offer some kind of insight into the game, or it's development that makes it interesting.

Pretty much every single system has complete game listings available on wikipedia, and you can almost always find some gameplay by googling the name, so it takes literally seconds to see if you'd be interested in a game or not.

>>3226619
>personally curated lists
The problem remains, that people are notoriously bad at curating them. You end up with the same 10 games over and over again, and it's ridiculously difficult to make flexible lists. You'd have to force people to do lists to all kinds of subjects, which just amplifies the problem

>it's inherently clickbaity
It is, which is why it's a good idea to not base it around rankings. The rankings are an emergent feature. You can introduce features like a "random good game", browse the games switching through criteria. Say, start with the top ten of the SNES, limit it to a single year, pick one of the games and get all the ones from the same genre, across all platforms, and so on. Browsing is a much more powerful way to discovery than lists, as the current mood has an impact on what you'll find, which no list can ever provide

>Pretty much every single system has complete game listings available on wikipedia, and you can almost always find some gameplay by googling the name, so it takes literally seconds to see if you'd be interested in a game or not.
Nice theory, but missing some big issues. Doing it that way, Wikipedia's lists are hard to cross-reference, it's difficult to order by criteria other than the ones in the table, and finding the gameplay infos and stuff is needlessly time consuming. The goal of such a site should be to perform this kind of aggregation, that an interested reader would be doing anyway, for them. Because with that chore over, the reader can focus on browsing, to get to where they want to be. Depending on how crazy a membership model should be, one could not only find a game, but also mark it as played or completed, which may or may not apply a different weight to their opinion on the next game-duel involving these games.

>>3226351
it's meant to be more of a search engine like you already know a type of game you're looking for. like "platformer" you may want to play. or specifically "metroidvania". if you wanna look for all famicom games for example so be it, but it's mostly meant as a "hey i'm looking a _______".

and to the other replies. it's more like i'm the main curator that listens to everyone as a whole. i'm actually an emulator developer who's run an online community for several years but i'll remain anonymous while doing this.

>>3226705
>like you already know a type of game you're looking for
So you start with the platformers list. You can still limit it by year, sub genre, publisher, etc. That was my suggestion. A "top 10" for every imaginable combination of search terms, thanks to a ranked DB backing it

>if you wanna look for all famicom games for example so be it
And then you filter that for platformers, or year, or developer, or ...
There is no difference between looking for one search criterion (genre) or another (platform), or any others.

>>3226712
yeah that's the thing with the hashtags. you can say #famicom #platformer and you're good if you wanna look for a famicom platformer.

>>3226726
again, why the # in front of them? And why not group them by category?

You're just describing an interface that performs these DB searches

>>3224589
>Guys, I've got a great idea
>You all work for me
>You do what I say
>And all the big decisions are mine because it's my idea
>READY GO

Have fun being your only unpaid intern, OP.

>>3226729
the hashtags themselves don't matter, it's just an easy way to keep it all as a single string in a database of sorts from a programming perspective. if i keep each like say... md5 checksum of a game title tagged with different hashes, it's much easier. i promise it makes more sense but in reality i can just make it "Super Famicom" on the front-end and then trim/lowercase it to "#superfamicom" and search the database for that string instead. it's really just a design choice. but i'm not averse to regular formatting.

>>3226737
i'm completely self-employed already doing game, app, and web work, dude. this is just for fun to give back to /vr/ and make it easier to recommend games to one another and discuss what we think should be on this site as a group.

you're a faggot.

>>3226740
>md5 checksum of a game title
or just the title itself, together with all title-specific information (release date, platform, genre, publisher, etc), like a normal rdbms

>single string
Are you insane? Well, have fun then

>to "#superfamicom" and search the database for that string
Why the #? What do you gain from it?

>>3226747
>or just the title itself, together with all title-specific information (release date, platform, genre, publisher, etc), like a normal rdbms
i just meant as an overall hash for its uniqueness. admittedly i can just do 0 to infinity.

>Are you insane? Well, have fun then
i meant a single string for each tag, not a single string for ALL tags.

>Why the #? What do you gain from it?
nothing it's hypothetical. but yeah we can just leave that out. i just like on similar sites like what.cd or passthepopcorn where they have tags like "punk.rock".

>>3226753
>i just meant as an overall hash for its uniqueness. admittedly i can just do 0 to infinity.
Most RDBMS have key generators for a reason.

>i just like on similar sites like what.cd or passthepopcorn where they have tags like "punk.rock".
Do you understand why some sites use a #? Do you understand why it's not useful for your case? Do you understand there there doesn't have to be a correlation between "tags" in the frontend and structured data in the backend?

If you do things like platform or publisher as tags, in the sense of strings associated with a game, you lose all kinds of meta information. Considering each game has a platform or a publisher, it's much more sane to have tables for platforms and publishers, and FK reference them in the game table. Tags in that sense of string <-> object are useful for unstructured data that lacks meta information. Even your example of punk.rock is probably somewhere internally marked as a genre, so the engine can distinguish it from other non-genre data.

>>3226767
>Do you understand why some sites use a #? Do you understand why it's not useful for your case? Do you understand there there doesn't have to be a correlation between "tags" in the frontend and structured data in the backend?
yes, yes, and yes. as i said it's just a hypothetical for familiarity for people used to the web >2.0 feel for searches and tags. i think you're taking it too literally when i'm not meaning to talk to other people with a programming background since we're on /vr/ and some of our fellow anons don't even emulate.

>If you do things like platform or publisher as tags, in the sense of strings associated with a game, you lose all kinds of meta information. Considering each game has a platform or a publisher, it's much more sane to have tables for platforms and publishers, and FK reference them in the game table. Tags in that sense of string <-> object are useful for unstructured data that lacks meta information. Even your example of punk.rock is probably somewhere internally marked as a genre, so the engine can distinguish it from other non-genre data.
agree to disagree on this one. you can keep platforms/genres in the overall tags and provide them on the front-end as a category if need be. it's just overall a search oriented idea from an amalgam regardless of platform/genre.

>>3226796
>the web >2.0 feel
that's a really powerful way to drive me away from a site. I'd rather navigate the content, than deal with the feel.

platform and genre were just two examples. You can expand this instantly with a ton of things. There is unstructured data that's useful for tags. But tags, on a conceptual level, are not a kill-all, and more often than not, vastly inferior to dedicated metadata.

>it's just overall a search oriented idea
Not really. Your hammer is a keyword search, and you're working hard to make sure your system is a nail. You can do that, of course, but I do believe it'll be far less useful than actual metadata. And again, what happens in the backend is completely unrelated from the frontend.

So your backend may have a platform table, with all platforms in your db, and a genre table with all genres in the db? You can still search for "nes and platformer". It's trivial for the backend to figure that one is a platform and the other a genre. Meanwhile you gain super valuable metadata. A simple question like "which system has the most platformers" is virtually impossible for a pure tag driven system. You'd have to determine all platform tags first, then you could perform a query. So how do you determine what's a platform tag? Some lookup table? Great, we're back to square one. Learn your data, model your data. What you see, do and search in the frontend does not have to reflect what is modeled in the database. What you model in the database though has great impact on your capabilities.

>>3226820
i'd keep them all as tags regardless and do the grouping of categories in the "front-end" of sorts. it'd be more of a "atari2600" AND "platformer" down through "famicom" AND "platformer" (or whatever is lowest at Z). all the tags are there regardless on a per-game&platform basis.

>>3226862
>do the grouping of categories in the "front-end" of sorts
declared in a front end file? That'd be a distributed data model, which is uncool. declared in a table? That's so close to an rdbms, yet not. Anyway, your site, your choices.
regardless, if you do get something to show, like a prototype, or a functional site, report back. I may not agree with all your decisions, but I strongly welcome people doing something to better gaming, or even just management of it.

>>3226872
thanks anonbro

>>3226743
I hate the recommend me threads as much as the next guy but there are literally dozens of sites out there dedicated to recommendations so:

Why is yours different?
How will it stop these threads from popping up?
How are you going to promote the site to get not only contributions, but also a dedicated userbase?

Until you can answer these questions you can just fack off ya faggot.

Programming is literally like half of a successful webapp.

>>3227484
>it's not good enough that you're spending your free time to do this for our benefit
>It has to do something for ME