STEAMPEEK
I love games: I admire the ideas, the creativity, the humor, the attention to details, also I respect the team, or the lonely individual behind a quality game. You would think that with those libraries, tools, tutorials, assets, templates, scripts, e-books it is much easier to build a game today than a few years ago - and you are right. Unfortunately it also means that it is much harder to find them - I mean the good ones and I don't like that feeling.
I have a constant whisper in my head that "you are missing a great game", "it is out there, you just don't know about it", "you'll probably never know it existed". Fear of missing out? Check. And I started to wonder what if I could do something about it, what if I could help myself and others discovering, finding and enjoying more games.
Well, it is easy to pump yourself up with energy and define a mission, but do something about it - well that is another story. I know that there are tools around which have features like searching for tags, and finding similar games, but, to be honest, I didn't like them. They were not even in the "good enough" category. I missed the wow effect, the moment when you find a new game, and say "wow, how come I never heard of this gem before". I believe a good recommendation system should give you "wow"s. Probably not always, and not every time, but it still should do it.
Ok, so recommending games. I know what results I don't like, and I can feel when a recommendation is not right. Can't tell you exactly what I feel, it is something like when you have a favorite meal, drink, snack and they change the recipe a bit - not much, probably just something is cheaper, and the boss with the suit says "change it, they would never taste the difference". (Yes, we do.)
This led me to the point, where I knew that I should try to build my own tool. I can code (don't laugh at the back, please) I have that mystic feeling (which should probably get a name by now), and I want it to work, I really do.
There are many algorithms out there for recommendations, and I'll be honest, for some I don't have a clue what they trying to do. With my algorithm I trust in my special feeling (which might be a disorder, who knows). What if I told you that coding isn't the hardest part here? Testing hundreds of results one by one, manually, by me - now that is tough. What works for an RPG, doesn't work for a casual? Retry. Works for strategy games, doesn't work for rhythm games? Retry.
One part of me was constantly trying to convince me, that "it is good enough", "even if there is one good result, it is worth to use it", "there are smarter teams working on this, what could one guy do about this". And the truth is, there is a point in every project, every goal, every work where if you just ignore these that is the straight way to insanity and/or depression (yeah, there is a slight chance that if you continue you will find and/or create something which would nobody appreciate, and after a couple of years you died, you will be called a "genius").
And here we are at a checkpoint, a milestone, a positive feeling about what is done, and how it works. A solid algorithm, a friendly tool, a fast system which is able to find similar games by a game, or a search by tags. It is a good feeling that it is done, and a better that I can share it with others, and with you - helping everyone finding something cool to play with.
Are we there yet? Is it done? You probably know the answer to these by now: I still have some plans, I will always find something to tweak, but currently I'm happy because I feel that I can help many gamers out there, who have the "Is there a game like..." question. (Also, if it helped someone making that whispering about missing out a little more quiet - that alone did worth it for me).