Chris Salisbury

I aspire to create and improve hardware, software, controls, and all things physical. I spend a lot of time on every other imaginable subject, too.

[Chris Salisbury @Chris_Salisbury]

@brouhaha Yeah, FTDI does have some good stuff if you can guarantee your supply chain and are able to pay the premium that requires. But FTDI-gate 1 was enough to boycott in my opinion, and ever since FTDI-gate 2, it's hard to even take them seriously anymore.

[Chris Salisbury @Chris_Salisbury]

@zipcpu As mentioned, any signal with significant discontinuities (c0 most notably, but c1 and c 2 as well depending on how their scale compares) would probably have similar benefits.

[Chris Salisbury @Chris_Salisbury]

@zipcpu I think there are several conveniences as well in terms of reconstructing (or just transforming really) data. I might not be understanding your point, but a frequency shift doesn't change the usefulness of wavelets; it's just a scaling operation, right?

[Chris Salisbury @Chris_Salisbury]

@trc_wm @zipcpu geographic information systems

[Chris Salisbury @Chris_Salisbury]

@trc_wm @zipcpu Computer vision - structure from motion / photogrammetry

[Chris Salisbury @Chris_Salisbury]

@zipcpu I suppose this could be due to discontinuities (or "non-uniform t/f resolution and/or fft windows longer than the transform") but those seem to be pretty compelling reasons o their own...

[Chris Salisbury @Chris_Salisbury]

@zipcpu I haven't kept up with new developments with wavelet deconstruction or new fft methods, but as of about 5 years ago wavelets were often more efficient in GIS and computer vision (specifically sfm and mvs) applications.

[Chris Salisbury @Chris_Salisbury]

@leviathanch @evacide "不是复杂" is not correct. Still awkward, but better would be "不复杂." Maybe you should reconsider if the aerial act is actually worse or not...

[Chris Salisbury @Chris_Salisbury]

@evacide I now want to hear that you solved the first nightmare by studying Chinese to the point where you are confident to pass any and all Chinese finals -- and that you are now practicing aerial acrobatics day and night. :-D

[Chris Salisbury @Chris_Salisbury]

@raphlinus Width is the only thing that causes me to switch away from Inconsolata on occasion, so this sounds extra super awesome. And plus a million for a font editor in Rust.

[Chris Salisbury @Chris_Salisbury]

@niko_yuwono お疲れ~!

[Chris Salisbury @Chris_Salisbury]

The Ghost of a Christmas Present https://t.co/xmukYBRPZW

[Chris Salisbury @Chris_Salisbury]

@wcswett Those both look like wonderful projects, thanks for sharing! If I'm ever in that area of the world again, I'll have to check back on them. Or maybe if I ever start another garden we can trade seeds. :-) For now I seem to be city-bound, though.

[Chris Salisbury @Chris_Salisbury]

@wcswett Are you the guy that did track-shun? Sorry if you aren't and that question therefore makes no sense. Feel free to ignore.

[Chris Salisbury @Chris_Salisbury]

@wcswett I'm the opposite -- I don't use Facebook much. But then I don't use Twitter much either. :-)

[Chris Salisbury @Chris_Salisbury]

@majek04 Maybe I'm just not reading enough qmail or openbsd code, but am I the only one that doesn't like the "goto" error handling style they use?

[Chris Salisbury @Chris_Salisbury]

@ablentry @Miss_Malware eXit ... :exit and :xit work, too, I think. Not only is it faster (which for me is the point of using vi), but it also only writes if there were changes -- so you don't have to think about whether the "w" will mess up your 'last modified' timestamp. :-)

[Chris Salisbury @Chris_Salisbury]

@sonotsecure I've never understood why so many people use :wq instead of :x

[Chris Salisbury @Chris_Salisbury]

@ablentry @Miss_Malware I've never understood why so many people use :wq instead of :x