Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
correa_brian
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
31.
▲
by
correa_brian
1y ago
seems like a lot of people in the comment disagree. good to know it exists
32.
▲
by
correa_brian
1y ago
:chefs-kiss
33.
▲
by
correa_brian
1y ago
Would have loved to use sqlite if I could get good results. Maybe I botched the implementation.
34.
▲
by
correa_brian
1y ago
Appreciate the feedback! We haven't setup the infrastructure for a trial but maybe in the future. That's cool you built a similar tool - what kept you from releasing it? Plan is to ship a Windows and Linux version in the next few
35.
▲
by
correa_brian
1y ago
why?
36.
▲
by
correa_brian
1y ago
For vector search we couldn't get my results to return meaningful entries. My preference would have been to use sqlite. Any others you recommend besides sqlite with the vss extension?
37.
▲
by
correa_brian
1y ago
How so?
38.
▲
by
correa_brian
1y ago
Haven't looked back since
39.
▲
by
correa_brian
1y ago
For now we're using CLIP. We've also done testing with Siglip and Gemma for a full-blown vision model.
40.
▲
by
correa_brian
1y ago
dont like the sound of reimplementing the whole UI components....
41.
▲
by
correa_brian
1y ago
I love Gemma. I use it on LM studio and am working on getting it into Desktop Docs. Thats for the nomic link. I'll do some testing...
42.
▲
by
correa_brian
1y ago
I've heard good things about Flutter! We briefly considered it but just opted for Tauri out of preference.
43.
▲
by
correa_brian
1y ago
The move from Electron helped us remove "native libraries" in Node. Previously we were using Xenova transformers to instantiate the models. This forced us to use multiple package.jsons and added a lot of cognitive overhead to our
44.
▲
by
correa_brian
1y ago
Great resources, thanks. I'll look into the other packages and compare against our onnx runtime setup.
45.
▲
by
correa_brian
1y ago
Nice. I'll try to test these out against our current implementation.
46.
▲
by
correa_brian
1y ago
I did some prototyping with SQLite + VSS extension to handle embeddings and wasn't getting the same results as using the Redis search modules. My preference would be to run something that doesn't require a separate for the DB, but
47.
▲
by
correa_brian
1y ago
Desktop Docs needs a GPU to work well, so we started with the Silicon chip Macs since they are ideal candidates to handle the workload of indexing an entire media library.
48.
▲
by
correa_brian
1y ago
What do you think would be more in line with what you've seen in the past? There's definitely room to keep optimizing.
49.
▲
by
correa_brian
1y ago
Yeah we didn't get a lot of traction on PH unfortunately and they have a lot of rules about relaunching. As far as users go - it ranges from prosumers to producers at media companies.
50.
▲
by
correa_brian
1y ago
For sure. We've been prototyping embedding a vision model in desktop docs, but just needs more dev time to be stable. Went with CLIP for parity, but we are looking to upgrade soon. I tried Siglip and wasn't impressed. Do you know
51.
▲
by
correa_brian
1y ago
That's it! Almost one year anniversary for Desktop Docs!
52.
▲
by
correa_brian
1y ago
Thanks for the feedback! Will get a demo video and contact info up shortly. De-duplicating images is on our roadmap. Shoot me your contact info at hello@desktopdocs.com. Would love to see if we can help.
53.
▲
by
correa_brian
1y ago
Should we do a c++ rebuild
54.
▲
by
correa_brian
1y ago
We considered it and decided all the wrangling wasn't worth it. Briefly also thought about adding in a python process to interact with our ML models, but figured it was worth streamlining this all into a single language/framework
55.
▲
by
correa_brian
1y ago
Awesome! Linux version is on our roadmap and that use case is exactly why we started building this too. I can let you know when that version is ready
56.
▲
by
correa_brian
1y ago
We haven't started testing for windows yet. Are you on Windows? Happy to let you know when we're releasing that version.
57.
▲
by
correa_brian
1y ago
After our Electron experience, I wish I had moved out of my comfort zone (JS) sooner. Electron just requires a lot of optimization and you have to be really tight with your imports to avoid loading things you don't immediately need. Th
58.
▲
by
correa_brian
1y ago
Thanks! If you're interested in that version, drop us a note: hello [at] desktopdocs dot com. We'll shoot you an update when it's ready!
59.
▲
by
correa_brian
1y ago
Yeah there are definitely hang ups along the way. It's a big of wack-a-model but thankfully we've had an easier time developing with Rust than on Electron.
60.
▲
by
correa_brian
1y ago
We aren't offering a trial right now, but to answer about docs --> we're testing an update to support searching text and will be releasing it soon. Right now you can search the contents of all your images and videos.
More ›