HN user

choochootrain

64 karma

[ my public key: https://keybase.io/choochootrain; my proof: https://keybase.io/choochootrain/sigs/6jElRqGhrkiXKMojYUbf1MeaesVxQlErPcHbldH2n-A ]

hurshal#imap.cc

Posts2
Comments32
View on HN

a few years ago i quit my full time job but negotiated a consulting arrangement with the company on an hourly basis.

it was beneficial for them because i was able to offboard and knowledge transfer much more effectively over a few months of part-time work as opposed to the standard 2 weeks notice.

it was beneficial for me because right out of the gate i had a consulting contract with a yc startup at a rate that was slightly higher than my yearly salary which i could use as social proof when reaching out to new clients and negotiating rates.

i ended up consulting for about a year as my only source of income where i took on a couple other clients through my network (ex-coworkers starting companies and needing help faster than they could hire, ex-coworkers referring me to others who were stating companies) and a few lucky connections through upwork (he contacted me on linkedin so upwork never got a cut of the invoices i sent). i've also tried toptal and moonlight but no contracts have come out of those platforms.

a year ago i took on a new full time job and i've been consulting on the side since. i agree with ananonymoususer in that i can have as much extra work as i can handle however i was lucky to have already created relationships with clients before getting a full time job. i'd imagine approaching a new client while maintaining a full time job is a harder sell than simply letting an existing client know that your availability will be reduced moving forward.

its a two-edged sword. if you can trust the application code to effectively schedule itself then it's good but you can also get bugs and perf issues where a synchronous chunk of code blocks progress on everything (see js blocking the main thread and causing the ui to freeze)

the pro is that your async code explicitly defines points where context switching is okay since you're blocking on something anyways. this could be good for perf if context switching in the middle of a synchronous operation is expensive.

the con is that your async code might not cede control often enough to allow other coroutines to make progress.

so yes, you can have something hogging the runtime but in the context of an application that you control as a whole this is something you can avoid/fix if necessary.

at the OS level this might not make sense because you have to assume that applications are adversarial and will try to hog cpu time...

what a blast from the past - i remember this game and it blew my mind when i first realized i had the power to plot a polynomial through all those points with little effort required

i think it's more about the culture which allowed that image (even in its cropped form) to proliferate as it did: > About six months later, a copy of the [Playboy] issue turned up at the University of Southern California’s Signal and Image Processing Institute, where Alexander Sawchuk and his team happened to be looking for a new photograph

i think we've come far enough to agree that having playboy issues in your research lab is no longer acceptable so why continue to implicitly endorse that?

find a new picture if you want. it's not like this is the optimal image for testing anyways...

i agree with you but you have to keep in mind that pages which load fast on a fiber connection and render and respond quickly on the latest macbook don't guarantee the same for the average user.

giving devs 4 year old laptops would help but when shiny startup is offering new macbooks how can you compete?

well then you must be inhuman because subconscious bias plays a part in online advertising. sure you can mentally block them but eventually you will fatigue from the cognitive overhead.

sorry to paint such a bleak picture but this stuff is ubiquitous, both online and offline, and it will eventually and consistently break your mental defenses in a way that you might not even notice.

i think the low contrast in your Geoff.Greer.fm logo and the "go dark" button might be what GP is talking about. i didn't realize the button had words until i saw the hover state had opacity: 1

i last used cloudformation over a year ago and it was a pretty shitty experience overall. bread and butter resources like vpcs would fail to provision in time which would cause a cascading failure and result in the entire stack failing and rolling back. also cleaning up stacks that failed to roll back was a bit of a pain when there were intermingled dependencies - the dashboard was about as unhelpful as possible in determining what order to delete things so i could carry on doing what i was actually trying to do.

i chalk that up to aws resources being backed by an eventually consistent db and the fact that aws has never built dashboards that are better than "acceptable".

it's possible that i'm missing something here but i've found terraform to be orders of magnitude better on both fronts. it takes a much more defensive stance on resource provisioning with retry logic and picking up on a previous `terraform apply` which timed out instead of rolling back the entire thing. and for whatever reason, terraform is usually way faster than cloudformation for provisioning the same resources.

i've been playing around with supercollider lately and i would love to see a writeup on your process for creating this. sound synthesis is really cool but i've been having trouble finding good resources for understanding things like ring filters at an intuitive level.

seems like it would go nicely with a procedurally generated game too!

in the past i have used cloud9 and nitrous.io for teaching, and i had students clone a template repo with exercises and lessons. this is awesome to me because its a much more structured way to manage projects, especially for students who haven't used git before.

one tool which i have built with my co-teachers in the past is a problem generator that tracks which concepts the user has "mastered" based on the number of problems they have solved for a given concept. we had a naive problem generator that used templated problem specs with randomly generated values: https://github.com/pftp/pftp-web/blob/master/practice/python... have you considered anything of the sort for students who want more practice?

i wonder how the medicare fraud strike force is currently doing this, and how we, as technologists can improve the process.

working around HIPAA makes this an particularly hairy problem but from what i understand it is still possible to create a compliant solution for hospitals, emr vendors, and insurance companies, and even patients to detect medical fraud.

i couldn't find much but maybe some work is already being done in this space?

DSP in Go 13 years ago

how do you recommend getting into this sort of stuff? i have some basic experience with microcontrollers and i'm interested in some dsp hardware projects but i'm not sure where to start as the classes offered do not touch hardware..

DSP in Go 13 years ago

I've never used Go before, but this is making me seriously consider trying it out. All of the DSP classes I've taken recommend Matlab + the DSP toolbox, and I would love to have a serious alternative.

[dead] 14 years ago

if youre bored with facebook and google and such, no one is forcing you to write about their every single move. there have been plenty of new and exciting things happening recently in medicine, transportation, energy, etc. they just don't happen to be the kind of things that techcrunch writes about...

Structure and Interpretation of Computer Programs http://mitpress.mit.edu/sicp/full-text/book/book.html

It is one of the classic CS textbooks still used by Berkeley (CS61A) and MIT (not sure) intro CS courses. Its not trivial stuff. While it wont teach you obscure data structures like ropes, it will expose you to a wide variety of topics including but not limited to: functional programming, lambda calculus, OOP, logic programming, client/server programming, non deterministic programming, streams as data, the meta-circular evaluator, lazy evaluation, and concurrency. I had a fairly strong CS background before taking CS61A at Berkeley, but this book (thanks to Scheme) taught me how beautiful computer science can be. Now working in Java is a complete turn off ;)