jmrko

6 ms·

1.
by jmrko 4mo ago
This is great, thank you! I predict agentic coding will change software dramatically in this direction. I recently had my weather app showing me a full-screen video ad. It drove me so mad that I developed wwads in a few hours with claude, W
2.
by jmrko 7mo ago
Have a look at this comment from Ken Hayworth, a highly respected scientist in the field, which I am copying here from a twitter post by him ( https://x.com/KennethHayworth/status/2032604687212392562 ). I also just
3.
by jmrko 1y ago
Great idea; I just picked IMAP because I thought it'd be somewhat universal. I hadn't heard of JMAP before. It should be quite straightforward to integrate. FastMail actually also seems to support IMAP.
4.
NeuralMail: Open-source tool to search all your emails with LLMs(github.com)
4 points by jmrko 1y ago|3 comments
5.
by jmrko 1y ago
I got tired of “email archaeology”; digging through years of inboxes to find something I know is in there. So I built NeuralMail: a fully open-source tool that makes your email (IMAP accounts) searchable with LLMs. It’s not an email client,
6.
by jmrko 6y ago
I don't think it is clear at this point that there are unexpected antibody titer drops in comparison to other viruses, Florian Krammer's work looks solid to me: https://threadreaderapp.com/thread/1285618977654
7.
Sprayable User Interfaces: Prototyping Large-Scale Interactive Surfaces(hcie.csail.mit.edu)
9 points by jmrko 6y ago|0 comments
8.
by jmrko 7y ago
That's a bad argument against DIY masks though. I tend to also agree with Zeynep Tufekci in this matter: https://www.nytimes.com/2020/03/17/opinion/coronavirus-face-...
9.
by jmrko 7y ago
I think when you remove ingroup and outgroup and just keep the argument for the attitude toward yourself/others it becomes more obvious what I mean. Otherwise it really depends on where you draw the boundary for the ingroup (e.g. your
10.
by jmrko 7y ago
Well, I think my optimum is just a hypothetical high-trust society in which everyone strives to be self-sufficient but gets supported in case of failure by others. Of course, this is not how the world works, which is more what you are descr
11.
by jmrko 7y ago
Very interesting comment, not because I agree with it, but I think it exposes a classic left vs right wing problem, and that the attitude is applied universally instead of selectively. IMO, the optimal attitude for society would be if every
12.
by jmrko 7y ago
Very interesting read. I interpret this in the way that clustering (eg HDBSCAN) on UMAP-projected data makes some sense at least (contrary to tSNE), are there any differing opinions on this? Interesting related discussions: https:/&#x
13.
by jmrko 7y ago
The recent NIH brain initiative report describes the acquisition of a synaptic resolution connectome of an entire mouse brain using massively parallel electron microscopy. The 5 year plan seems a bit too optimistic to me, especially due to
14.
Brain initiative: Connectome of an entire mouse brain (human in decades?) [pdf](acd.od.nih.gov)
1 points by jmrko 7y ago|1 comments
15.
by jmrko 8y ago
I just signed up for this, maybe that's the reason?
16.
by jmrko 8y ago
Accessible review articles: https://dspace.mit.edu/handle/1721.1/96185#files-area https://www.sciencedirect.com/science/article/pii/S095943881... Sebastian Seung's book: http:
17.
by jmrko 8y ago
That's correct - I imaged the volume and we extracted the synapses before ( https://www.nature.com/articles/nmeth.4206 ).
18.
by jmrko 8y ago
One of the authors here; Imaging synapses is really no problem with electron microscopy resolution. In fact, you can see automatically extracted synapses (little spheres) in the video of the blog post, this was done a year earlier already!