jgrahamc

73 ms·

811.
by jgrahamc 3y ago
Email me and tell me jgc@cloudflare.com
812.
How to beat an adaptive/Bayesian spam filter (2004)(blog.jgc.org)
3 points by jgrahamc 3y ago|0 comments
813.
by jgrahamc 3y ago
See: https://blog.cloudflare.com/privacy-first-web-analytics/
814.
by jgrahamc 3y ago
Just to be clear: there's no Cloudflare special case here. We're not sending you to 1.1.1.7. We just send whatever the archive.is auth servers decide to send to us. They are auth after all. If we returned 1.1.1.7 it's because
815.
by jgrahamc 3y ago
https://blog.jgc.org/
816.
by jgrahamc 3y ago
Yes.
817.
by jgrahamc 3y ago
Good things come to those who wait :-)
818.
by jgrahamc 3y ago
No. It’s running on the same machine. Our goal is distributed ML/AI in line with our usual mechanism of running everything, everywhere. That is powerful because it means anything we build scales with our network.
819.
by jgrahamc 3y ago
No. That’s wrong. We don’t “manage regions”. 300 data centers. Everything running everything. Only time you’d need to think of a region is if you have some compliance situation (“keep all my data in North America” or “only terminate TLS in
820.
by jgrahamc 3y ago
No.
821.
by jgrahamc 3y ago
Cloudflare does not have "regions": your stuff runs everywhere on our network. And we can be really smart about where's the best place: https://blog.cloudflare.com/announcing-workers-smart-placeme...
822.
Lost in transit: debugging dropped packets from negative header lengths(blog.cloudflare.com)
2 points by jgrahamc 3y ago|0 comments
823.
The Cloudflare Radar Internet Quality Page(blog.cloudflare.com)
13 points by jgrahamc 3y ago|0 comments
824.
A step-by-step guide to transferring domains to Cloudflare(blog.cloudflare.com)
2 points by jgrahamc 3y ago|1 comments
825.
by jgrahamc 3y ago
Enjoy: https://blog.jgc.org/2013/05/from-ground-up.html
826.
by jgrahamc 3y ago
Yes. I was thinking about that after my comment. All the folks doing machine learning optimizations, weird stuff with AVX2, etc. are the same folks who would have been stuffing chess in 4K of ROM and 128 bytes of RAM. In 1985, me and anothe
827.
by jgrahamc 3y ago
Anyone who wanted to push the limits of hardware at that time had to do tricks like this. There just wasn't any other choice. Sure, there were plenty of programmers back then who weren't capable. But it's not magic, it's
828.
by jgrahamc 3y ago
If you only had 4K of ROM and 128 bytes of RAM you’d find a way. That’s what we all did back then. This isn’t to diminish Atari’s achievement (this stuff was super hard and also super satisfying), it’s just to say that doing these kinds of
829.
by jgrahamc 3y ago
Are they Model 5150s?
830.
INP. Get Ready for the New Core Web Vital(blog.cloudflare.com)
1 points by jgrahamc 3y ago|0 comments
831.
HTTP/3 Prioritization(blog.cloudflare.com)
31 points by jgrahamc 3y ago|1 comments
832.
by jgrahamc 3y ago
Wonderful. Considerably more complex than getting a Minitel to speak to a Linux box: https://blog.jgc.org/2022/03/using-minitel-1b-as-serial-term...
833.
How Cloudflare runs machine learning inference in microseconds(blog.cloudflare.com)
6 points by jgrahamc 3y ago|0 comments
834.
How Orpheus automatically routes around bad Internet weather(blog.cloudflare.com)
3 points by jgrahamc 3y ago|0 comments
835.
Every request, every microsecond: scalable machine learning at Cloudflare(blog.cloudflare.com)
8 points by jgrahamc 3y ago|1 comments
836.
by jgrahamc 3y ago
It's information that's only really useful to us. It refers to a set of machines running the same version of our software. Part of how we do progressive rollouts of software.
837.
MiniMit – Bringing Minitel back to life(xn--multipli-i1a.fr)
50 points by jgrahamc 3y ago|15 comments
838.
by jgrahamc 3y ago
Way more than that now.
839.
by jgrahamc 3y ago
No. Not at all. Watching this feels like a total waste of time.
840.
by jgrahamc 3y ago
I don't want to be marketed to, what I want is other developers to recommend a product to me.

More ›