HNHacker News
TopNewBestAskShowJobs

sansnomme

1,145 karma · joined January 22, 2019

submissionscomments
sansnomme··on Ask HN: What Technologies to Learn in 2020?
(Assuming a CS background) To do anything useful/fun/interesting in bio you should have a strong understanding of the Central Dogma, once you understand that you can move on to the rest. Many here recommend building gene analytics and other similar software/SaaS. I don't recommend it because you learn absolutely nothing from those low hanging fruits. Genetics and its relation to CS, at a sufficiently low level, is mostly string manipulation and search. There is a market if you are willing to build and do sales but it's hardly exciting. Better to get some actual wet lab experience and understanding, than become yet another "data science" biotech startup. We have had enough of those in healthcare already, tons of so-called "health tech" companies that were merely performing analytics on wearables. Profitable? Maybe. Exciting and innovative? No.

Some reading material:

Synthetic biology: A Primer

An introduction to systems biology (get the 2020 edition)

O'Reilly: Biobuilder

Molecular biology of the Gene

Campbell's Biology

YouTube Channels:

The Thought Emporium

Josiah Zayner

Khan Academy Biology

Biology Professor

Shomus Biology

(The first two deal exclusively with bioengineering)

sansnomme··on Show HN: Dark Mode Hacker News Reader That Shows Top Comment
Congrats on shipping! If possible, add an option filter out "is Hiring" posts.
sansnomme··on Synthetic Modelling of Biological Communication [pdf]
Uri Alon just released the second edition of his classic text on systems/synthetic bio. Recommend checking it out.
sansnomme··on Tabula Scripta: A spreadsheet for small data-driven apps
Instead of your own language, I suggest using the Red language. It's similar enough and has a history of being used in RAD/spreadsheet software.
sansnomme··on Heliogen, backed by Gates, says it has achieved a solar breakthrough
Ssh, hacker news hates simple solutions like fiducial markers. I wonder if HN would dare to ride a spaceship that uses neural network to perform docking alignment.
sansnomme··on Show HN: Neural networks tutorial series with code
How else do you think AMZN/GOOG/MSFT are able to get you to use their turnkey ML-as-as-service?
sansnomme··on Show HN: Tailwind Button Playground
Think React-style. A button becomes a component. Otherwise your page may look noisy with each element having 5-10 tags attached.

https://tailwindcss.com/docs/extracting-components/

You can either use their built-in system or whatever your JS framework provides or both.

sansnomme··on Show HN: Tailwind Button Playground
It's a lot more maintainable than raw CSS but when you grow big you will have to refactor. The main thing about Tailwind is that the author is a very, very talented designer who also produces high quality YouTube tutorials. It's very easy to make Stripe-style landing page with Tailwind because all the examples are in that style. Most of your existing frameworks like Bootstrap, Bulma may look unique but when you want to ship a pretty page you will need severe customisation together with a fair bit of design talent. At the end of the day, you want your site to look beautiful, not just yet another wannabe-unique-glitzy-slightly-modified-themeforest site.

https://youtube.com/channel/UCy1H38XrN7hi7wHSClfXPqQ

His videos are worth their weight in gold even if you don't use Tailwind. From Figma design all the way to pixel perfection. You won't be able to find this level of expert knowledge on Udacity.

sansnomme··on Show HN: Constraint-Based School Timetabling with Prolog
Thank you for your hard work in promoting Prolog! It has made a great impact on the community. Do you think you can do a few posts on interfacing Prolog with external systems? E.g. not just manual, carefully crafted proof-of-concept demos. But a full interface with perhaps an API endpoint or database with utf-8 data. A lot of Prolog tutorials refuse to acknowledge the fact that it has poor support for non-proprietary Web/GUIs, compared to at least JS/Ruby/Python. It would be really helpful if you can do a post illustrating how to get from the step of multi-language data entry in e.g. React with a Python backend to getting it processed in Prolog.
sansnomme··on The Economics of Building Knowledge Bases
Also for certain products it's almost impossible to sell at scale without an enterprise sales team. Developers won't be "willing" to pay or admit the need for what's basically a glorified web scraper concept that has existed for as long as the web itself. For this sort of product to be successful you need enterprise sales, tons of funding and marketing, among other stuff. Even if your actual product is terrible you will still have a larger segment of market compared to a competitor product that may be more advanced functionality-wise. This is not something that's popular on HN because of the constant myth that "if you build a well-engineered product, customers would magically show up".
sansnomme··on Squoosh: Make images smaller using best-in-class codecs, right in the browser
Do not forget parallelism. Neural networks, electronic circuits, and a lot of other processes in nature occur simultaneously at an enormous scale.
sansnomme··on Show HN: Chaskiq – an open source alternative to Intercom, Drift and others
Well done! Congrats on shipping, you guys should meet up with folks from https://github.com/chatwoot/chatwoot
sansnomme··on Don't Shave That Yak (2005)
This is actually part of the business justification for using SaaS/PaaS instead of rolling your own. E.g. skip implementing auth when just starting and simply use Firebase.
sansnomme··on Happy New Year HN!
Happy New Year HN! Here's to fun, profit, and happiness!!
sansnomme··on I made my own image sensor and digital camera [video]
Certain FLIR resolutions beyond a certain density are, let's just say, not unlike the early days of cryptography.
sansnomme··on Reasons to Learn Redux as a JavaScript Developer
Yes it is pretty straightforward though the project is still growing and suffering from the usual growing pains. Think docs that has plenty of room for improvement, insufficient example projects and components. On the up side the team responds quickly, the concepts are ridiculously simple compared to Redux and there's a burgeoning community around it.

E.g. https://sveltematerialui.com/

The Next.js equivalent for SSR is Sapper: https://sapper.svelte.dev/

Note the creator of Svelte is the same guy who created the Rollup bundler so pretty much everything comes with dual configs for both Rollup and Webpack.

sansnomme··on Reasons to Learn Redux as a JavaScript Developer
Check out Svelte. Essentially reactivity baked into the compiler so you don't have to use observables everywhere.
sansnomme··on Reasons to Learn Redux as a JavaScript Developer
Redux is tremendously verbose though. Message passing isn't attractive when the state space is big.
sansnomme··on Show HN: A tiny isometric city builder in JavaScript
Would you mind adding support for touch devices?
sansnomme··on Max Crashes Strengthen Resolve of Boeing to Automate Flight
There are tons of math majors produced every year. Would be a viable business model to offer factory-style program-correctness-proving-as-a-service. Current proof services are mostly niche static analysis for stuff like Ethereum contracts and occasionally a wannabe legal-tech company. I think aerospace could use more proofs. You can't use Ada for everything, eventually you will have to code in C/C++.
sansnomme··on Squoosh: Make images smaller using best-in-class codecs, right in the browser
Kilo base pairs among other things.
sansnomme··on Show HN: The app that pays you for saving money through a monthly lottery
Do users still get the usual interest rates?
sansnomme··on Show HN: Tailwind Starter Kit – Free and Open Source UI for TailwindCSS
Any plans for Svelte? Also there is an overflow issue on mobile (popovers page). The landing page rocket icon on mobile needs additional margin on top.
sansnomme··on Show HN: Keras-OCR is an end-to-end, trainable OCR pipeline
You should market this more, there is a severe need for a Tesseract replacement, it has been falling behind current state of the art, especially compared to many cloud offerings.
sansnomme··on McDonald's holds communities together (2016)
Most countries' universities are not "world class". They are good but the top US universities are like tech unicorns, entirely in their own class when it comes to innovation/rankings, network, and opportunities.
sansnomme··on McDonald's holds communities together (2016)
Ah yes, the classic journalistic arrogance. If you see a bloated child keeled over in a third world country, do you simply photograph or do you help? Perhaps it is a failing of humanities education that this sort of issues are still worthy of debate, that the basic human instinct to give aid to a fellow man can be supplanted by philosophy and economic debate. A perfect example of why tertiary education is under attack from all sides; if the ivory towers brings marginal value and charges tremendously high rent, is it still worth keeping them around? A good topic for HN to consider and think about, it is easy to bash academia and universities as overpriced and irrelevant, but Khan Academy doesn't exactly fulfill the niche either.
sansnomme··on A Desperate Plea for a Free Software Alternative to Aspera (2018)
If they are like Riot and have their own fiber, it is not unlikely. There are tons of boring optimizations open source software also won't bother with without significant sponsorship (some could be specific to a particular file format). If you look at most network (and also video) protocols, they are accumulations of decades of improvements rather than neural network style overnight breakthroughs.
sansnomme··on Stroom – a scalable data storage, processing and analysis platform
There are a bunch of static analysis companies in crypto-contracts. Infosec could be a good pivot if the bubble pops.
sansnomme··on Is College Still Worth It? The New Calculus of Falling Returns [pdf]
I will just link to this.

https://philip.greenspun.com/bg/

sansnomme··on Computer Science from the Bottom Up (2016)
Linear algebra, discrete math, calculus. The mathematical underpinnings of algorithms and data structures. Turing machines will be a lot more useful once you actually know how to use properties to prove more things. A kindergartner can recite the layman description of an infinitely long tape with ones and zeroes but such simplistic understanding has very little practical use if you don't understand how it fits into the context of CS in general. Machines improve, architectures evolve, frameworks change. But math doesn't. Every few days there is a new "linear algebra for machine learning" guide that pops up on HN and every now and then there will be a "how to learn math" question on HN. The lack of mathematical maturity among software devs and engineers is not a good thing and reflects poorly upon the industry. Too many universities these days focus mostly on practical leet coding rather than the theoretical underpinnings of CS. An in-depth study of the finer aspects of a networking protocol would become outdated the moment the next iteration comes out, but a close examination of Shannon and information theory will serve you well for life. There seems to be a continuous myth that undergraduates cannot code a FizzBuzz to save their lives and thus all focus should be placed on testing that specific skill. This mentality of being dismissive towards math is pernicious to computing as a science and relegating theory as "stuff I never ever needed in my n years in industry" creates a harmful echo chamber for software engineering.
← PreviousPage 6 of 21Next →