Open DavidJavet opened 1 year ago
We never code the randomness, I just re-read the code to be sure and we never integrate this feature :)
I wanted to reach out to discuss a challenge I've encountered while working on our randomness algorithm.
So, here's the scoop: I've been diving into implementing randomness, and I've hit a bit of a roadblock. It turns out that having games in a random order on each page might result in duplicates popping up across multiple pages (because of the randomness not being aware of results from previous pages). Not ideal, right?
I've been thinking over a potential solution involving a Random Seed (https://github.com/elastic/elasticsearch/issues/1170) approach to maintain consistency for users throughout their session. But, I'll be honest, it's quite a hefty task, especially on the frontend side. Given the current bandwidth of @tonifisler and considering my own lake of skills on the frontend technology used, diving into this might be a bit ambitious right now.
I'd love to get your thoughts on how we should proceed: @DavidJavet @yrochat ?
Thanks a bunch for your time and insights! @DavidJavet @yrochat
Hello Kevin, Thanks for this. I see the issue and it's a good thing we reconsider adding randomness if it means so much work and uncertainty of results. Whatsmore, everytime there's a "shuffle" I am massively worried one of our "porn" Swiss Games end up on the first results of the Homepage. So as long as we have not solved this problem (maybe by censoring the first image of each porn game), I might be better not to implement randomisation.
I'd love to get your thoughts on how we should proceed: @DavidJavet @yrochat ?
- Should we push ahead with randomness on the homepage (minus filters), knowing that pagination might take a hit? I would say NO.
- Don't go further with Randomness on Homepage I would say YES for the time being, especially if it means lots of work on your side.
- Unlocking budget and implement this kind of heavy complex feature If it is such a heavy complex feature, yes, it should be something someone getting paid should do.
Thank you @WengerK for having a look at this and both you and @DavidJavet for the discussion.
I think option 3 – Unlocking budget and implement this kind of heavy complex feature – is the best, especially since we recently gained more recognition for the work done when the National Library decided to archive our work.
I have too many such funding demands in the queue so I won't make any promise but we could in the future ask for a small grant that would allow us to develop a list of features on the frontend (and backend if help is needed there too).
There's always the same selection of games appearing on the Homepage