eichin

As far as I can tell, the reason for the lack of companies is that they've made it impossible to monetize "software for other people" and apparently vast swathes of what used to need management are now "I can just build the parts of that I care about, it doesn't have to be good enough to ship to anyone else." (This horrifies me, though there is a silver lining: the value of the "idea guys" has also gone to zero :-)

So negative that the first I heard about this was with the title "New York Times adds glue to its Cooking App. Users aren't thrilled" (presumably from someone who read the niemanlab article before I did)

At first glance I almost misread this as a different C crime, since the declarations aren't in scope... uint8_t dhm_p[2048]; and uint8_t *dhm_p = (...)malloc(2048); both make dhm_p a reasonable pointer to 2k bytes of storage and are totally suitable for that argument, but sizeof the first one is 2048 and sizeof the second one is 4 (or more likely 8 these days, but this is specifically code from an embedded web server so I can say 4 :-)

But no, the actual problem was so much less subtle.

Nice thing about this one is that we have the implementation details right in the article:

    mbedtls_ssl_conf_dh_param_bin(conf, dhm_p, sizeof(dhm_g), dhm_g, sizeof(dhm_g));

The first sizeof(dhm_g) should have been sizeof(dhm_p). So, do you attribute this to a simple typo (hard to catch at a glance due to similar letter shapes), "single letter variable names are bad" but the wikipedia page on DH also calls them p and g - it takes a little digging to get that "root" and "modulus" would have been better names, but maybe having them directly match the most common form of the equation would have made them easier to review...

Or do you just add it to the list of C crimes, that "passing storage length and pointer" is maybe the worst possible way to return written-to storage from a function :-)

Same - it gets the ludicrous reach by having a small sensor (and OK but not miraculous lenses) but it turns out that most birdwatching is done in bright sunlight, which makes up for the optical limitations. (I also find it's easier to look for birds naked-eye first - even with mild corrective lenses, vastly more field of view and more effective motion-detecting reflexes - rather than trying to scan with the tiny field of view of the camera. But part of that was noticing the feeling of "levelling up" when I first started bird watching specifically :-)

There were a bunch of high performance computing efforts from the 1990s and on where, if you could work out that you'd need 5 years of "today's" compute, you should obviously hang on to the money for 4 years, then spend it on the correspondingly improved hardware and finish the job in a under a year. I've been considering building a box to run OLMO (one of the only models with training data "that you wouldn't have to defend in court" as far as I can tell?) and even with Sam Altman's supply chain attack, stalling for bigger hardware still seems like the right path...

Also, that particular product was disappointing compared to everything else they made - there was a full split keyboard, a half keyboard (about the size of the iGesture pad but had at least 5-finger multi touch as well as being combined keyboards and touchpads - the same "win" as the thinkpad or j-mouse, you can do gestural stuff without your hands leaving the keyboard.) Still had a lot of learning curve but they had broad sweeping gestures for open/close file and the like - though the one really odd one that I remember was "touch all four corners simultaneously" which would cause the keyboard to "type" about 80 characters of product and firmware version and "mode" information. (At the time all of the gestural mapping was in the keyboard itself, no drivers needed - and you could toggle some of the general "editing" gestures among windows/mac/emacs/vi presets.)

The emphasis on using optimus leg parts for actuators is interesting, given that they're also still under development...

(Ah, the youtube "designer interview" says around 1700kg (the first-gen was 1300kg; formula one has a minimum weight of 800kg.)

If you want that, Terrafugia has been around for a while - but you still need to drive it to an airport to deply the wings and take off. (Flies for rather more than 3.5 seconds, too :-)

Didn't F1 cars briefly have downforce fans (basically suction fans underneath the car) until it was ruled out of spec? Basically a "reverse hovercraft" trick, normally used for handling, but "turned up to 11" here.

The page talks about a bunch of weight reduction tech, but I didn't see an actual final weight for the Rocket model? Given that it can support itself with "downforce" (and no thrusters, and apparently not much momentum?) at the top of the loop it's got to be pretty light...

The joke "God is Real (unless declared integer)" is definitely fortran-specific, though :-)

Heh, I used that TRS-80 (microsoft!) Fortran in high school - wrote a library for propagating error bounds and significant figures through calculations, for chemistry lab reports. (Could probably have done it in BASIC too, but I suspect I was looking for an excuse to use fortran for something :-)