I haven't tried those features but I did try Meilisearch awhile back and I found Typesense to index much faster (which was a bottleneck for my particular use case) and also have many more features to control search/ranking. Although just to say, my use case was not typical for search and I'm sure Meilisearch has come a long way since then, so this is not to speak poorly of Meilisearch, just that Typesense is another great option.
Meilisearch just improved the indexing speed and simplified the update path. We released v1.12 and highly improved indexing speed [1]. We improved the upgrade path with the dumpless upgrade feature [2].
The main advantage of Meilisearch is that the content is written to disk. Rebooting an instance is instant, and that's quite useful when booting from a snapshot or upgrading to a smaller or larger machine. We think disk-first is a great approach as the user doesn't fear reindexing when restarting the program.
That's where Meilisearch's dumpless upgrade is excellent: all the content you've previously indexed is still written to disk and slightly modified to be compatible with the latest engine version. This differs from Typesense, where upgrades necessitate reindexing the documents in memory. I don't know about embeddings. Do you have to query OpenAI again when upgrading? Meilisearch keeps the embeddings on disk to avoid costs and remove the indexing time.
Thank you for the response here. Not being able to upgrade the machine without completely re-indexing has actually become a huge issue for me. My use case is that I need to upgrade the machine to perform a big indexing operation that happens all at once and then after that reduce the machine resources. Typesense has future plans to persist the index to disk but it's not on the road map yet. And with the indexing improvements, Meilisearch may be a viable option for my use case now. I'll be checking this out!
I hate the way typesense are doing their « hybrid search ». It’s called fusion search and the idea is that you have no idea of how well the semantic and full text search are being doing, so you’re going to randomly mix them together without looking at all at the results both searches are returning.
I tried to explain them in an issue that in this state it was pretty much useless because you would always have one or the other search strategy that would give you awful results, but they basically said « some other engine are doing that as well so we won’t try to improve it » + a ton a justification instead of just admitting that this strategy is bad.
We generally tend to engage in in-depth conversations with our users.
But in this case, when you opened the GitHub issue, we noticed that you’re part of the Meilisearch team, so we didn’t want to spend too much time explaining something in-depth to someone who was just doing competitive research, when we could have instead spent that time helping other Typesense users. Which is why the response to you might have seemed brief.
For what it’s worth, the approach used in Typesense is called Reciprocal Rank Fusion (RRF) and it’s a well researched topic that has a bunch of academic papers published on it. So it’s best to read those papers to understand the tradeoffs involved.
> But in this case, when you opened the GitHub issue, we noticed that you’re part of the Meilisearch team, so we didn’t want to spend too much time explaining something in-depth to someone who was just doing competitive research, when we could have instead spent that time helping other Typesense users. Which is why the response to you might have seemed brief.
Well, in this case I was just trying to be a normal user that want the best relevancy possible and couldn’t find a solution.
But the reason why I couldn’t find it was not because you didn’t want to spend more time on my case, it was because typesense provide no solution to this problem.
> it’s a well researched topic that has a bunch of academic papers published on it. So it’s best to read those papers to understand the tradeoffs involved.
Yeah, cool or in other word « it’s bad, we know it and we can’t help you, but it’s the state of the art, you should instruct yourself ».
But guess what, meilisearch may need some fine-tuning around your model etc, but in the end it gives you the tool to make a proper hybrid search that knows the quality of the results before mixing them.
I think this is a good example of why people should disclose their background when commenting on competing products/projects. Even if the intentions were sound, which seems to be the case here, upfront disclosure would have given the conversation more weight and meaning.
+1 typesense is really fast. the only drawback is starting up is slow when index getting larger. the good thing is full text search (excl vector) is relatively stable feature set, so if your use case is just FTS, you won't need to restart very often for version upgrade.
On one hand you have a potentially infinite variety of people and on the other you have a practice that works for just some of those people. And every practice that works for just some people, there are others taking advantage of that fact by saying, "Hey this works for everyone", being generally scammy to get people to sign up for something and make money. After all, capitalism justifies and awards that. But outside of economic motivation, on a much more personal level, the practice might actually work REALLY well for SOME people. And this applies to a lot of things.
When it comes to mindfulness, in my experience, for SOME people, their mind is so far overthinking things. About themselves, about the people around them, every possible factor in life is fair game really. This can end up causing anxiety and impulses to take control of actions. It's possible to have so much physical tension built up all around the body all the time that it feels shut down. In these cases, learning how to just close your eyes and let go of all those thoughts and all that tension... in your face, in your shoulders, around your hips... bringing your mind away from those thoughts and that tension by focusing on your slowwww breath, might just bring you to a place of comfort that you didn't realize was possible. In the end what all of this represents is stress. For someone who has stress and benefits from this practice, it's an actual battle to let go of the thoughts and tension. But if you get there, you have a whole new level of growth you can do from there. In that case, this practice feels like a super power. So you might want to let everyone know. But if none of this applies to you, it will probably look like a scam.
This is a great idea. I can imagine lots of parents finding this site through Google and making good use of it. Some SEO work may be worth it to make this easier for people to find. The one page dynamic nature might deter this but some simple workarounds would fix that.
Thought I'd note, however, that when I changed the district to Los Angeles and hit See Results. It remained in an endless loading screen and would not return results.
Even if you know where Pliny the Younger is being served, there's still no easy way to do this. They release a limited amount and it usually sells out in the first hour or so. If you go to the Russian River brewery it's a 5 hour+ wait. Crazy stuff.
It can be annoying to have multiple "layers" of volume control: one for the player, one for the TV, one for the receiver/sound system, etc. It's arguably a lesser evil to leave volume completely up to the user's TV or audio system, the way Apple TV does.
But then you need multiple remotes, which I feel is the greatest evil of all. I agree the Roku doesn't need its own volume control, but if you stick one on there, you can make it programmable to control the volume of your TV.
That would be ideal, though it might complicate the experience for non-techies, especially during setup.
Tangent: one thing that drives me nuts about universal remotes that they are all modal; ie, you must put the remote in receiver mode before you change the volume, etc. I wish I had a remote that could remember device presets for each button simultaneously, such that volume always uses the receiver RF, D-pad always uses Apple TV RF, etc.
If you took that introduction so deeply personally that you stopped reading out of offense, I think you have deeper issues to deal with than the fact that you're not taking advantage of this open source library. This is very much your issue, not the author who has just built a library and then organized documentation just to offer it all for free. I believe he should include whatever sense of humor he wishes.
I really hope that open source authors continue to include their character in their projects. It makes the world a more interesting place.
I haven't tried those features but I did try Meilisearch awhile back and I found Typesense to index much faster (which was a bottleneck for my particular use case) and also have many more features to control search/ranking. Although just to say, my use case was not typical for search and I'm sure Meilisearch has come a long way since then, so this is not to speak poorly of Meilisearch, just that Typesense is another great option.