314 karma · joined April 21, 2015
Email: ian@chesal.net
[ my public key: https://keybase.io/irc; my proof: https://keybase.io/irc/sigs/r-ZwLHIHt_j1NxILTDpgiT_vEiH8xdtrLZZFxKcYA5Q ]
git has its place but it's really broken the world for how to think about SCC. There are other ways to approach it that aren't the ways git approaches it.
Perforce handled it all like a champ.
People who think large files don't belong in SCC are...wrong.
I'll enjoy my last year with GM in 2024 I guess.
Of the options I have for BOS <> SFO, Delta still probably has the best hard product offering there. But this is definitely changing my dogged loyalty. I'll be trying Alaska soon enough. Not sure I'll ever do United again...
It'll be a while before we see highly reliable and well-tested dual-reservoir systems is my bet.
The complexities of balancing insulin and glucagon in a two-pump system are also high. And the feedback loop from sensors that detect BGL aren't super fast. My kid's Dexcom works on a 5 minute sample loop now. So you can't make decisions fast and when you do, you can't course correct a bad decision quickly.
I love that people are working on this stuff. The folks at https://wearenotwaiting.net/ are amazing and we even use NightScout here, but the fragility of the systems are stark and it'll be a ways to go before it's not just the brave pioneers pushing these frontiers for T1Ds.
All the well-tested pumps (Tandem, Omnipod, Medtronic) are insulin-delivery only.
It was a little buggy when Ventura dropped, but it gets frequent updates and has stabilized in the past few months.
Notes here: https://gist.github.com/ianchesal/5c96c566ab99b60c2c557711fa...
Horses for courses I guess.
Bike looks neat though.
https://gist.github.com/ianchesal/5c96c566ab99b60c2c557711fa...
I thought they were in the other comment. The migration is pretty simple now that heroku only hosts the static part of the site and the data is in Atlas.
I mentioned the guide and my work on the NightScout discord. If there’s interest I could do from scratch instructions for those doing entirely new setups.
Note: I’m the person who did the migration and wrote the notes linked to above.
You can’t offer engineering services to the public without being licensed.
Non-PEO licensed software engineers exist in large quantities in Canada without any issues. They provide services to a company, not the public.
Diabetes is a shit autoimmune disease. But the tech gives us some hope. And I treasure every day, no matter how hard, I get with my kid.
Hang in there. It gets easier to manage as they get older and more involved in their care.
Not only did you get the charts, but you got their notes. And the notes of anyone before them from the book’s lineage.
Still have. Still use it.
It was definitey a clear delineation between _computer science_ and _engineering science_ during my undergraduate courses.
Japan was their first international foray. There were visions of the serene design aesthetics of Square clearing the clutter from countless Japanese shops. It was going to be the perfect society to embrace Square’s vision and form for credit card based payments.
Guess who has a banking system that’s archaic and runs on faxes? And a society that largely eschews credit and credit cards?
It was not the glorious meeting of culture and technology they had envisioned.
The pre-launch research was poor. It was a vanity expansion.
Free tier users, maybe not.
But paying users care very much about what they can expect for their outlay of money in the years to come. Roadmap presentations for paid software are _de rigueur_; they help keep your paying customers on your platform.. They're just not often shared openly like this.
Github has a vested interest in attracting more and more free-tier customers though. Then they go to companies and influence tool purchasing decisions and help feed the paying customer funnel. So this seems like a risk with some postive calculus behind it. It could be that free-tier users didn't know they wanted to know the roadmap, and now that they do they'll be happy they know about it.
...and was then asked at every turn, by every website and application and whatnot, to provide four or more digits of this number to accomplish even the most benign things.
There's what the US government thought it would be and then there's what it's become because zero enforcement on use of it as a national person identity number was ever enacted.
The "don't give anyone your SSN" trope has become one of those household jokes. Right up there with "American's don't pay high taxes".
If you want to dig deep into computer science history, you’ll note that LSF was once a slightly modified Condor. There’s a wild history between the two and U.Wisc and U of T.
I was routinely using Condor in semi-conductor R&D and running batches of jobs where each job was running for many days -- that's probably far longer than any single instance of a service exists at Google in this day and age, right?
None of the batch stuff does the networking management though. No port mapping, no service discovery registration, no load balancer integration, etc. That's Kubernetes sugar they lack. But...has never struck me as overly hard to add, especially if you use Condor's Docker runner facilities.
Edit: I should say that I don't _really_ think you could swap out Kubernetes for Condor. Not easily. But it's always been in my long list of weekend projects to see what running an cluster of online services would be like on Condor. I don't think it'd be awful or all that hard.
The other killer Condor tech is their database technology. the double-evaluation approach of ClassAd is so fantastic for non-homogenous environments. Where loads have needs and computational nodes have needs and everyone can end up mostly happy.