Would love the ability to search through these materials that are audio only.
832 karma · joined January 30, 2011
Would love the ability to search through these materials that are audio only.
Tried biking instead, but indoor biking is the most mind-numbing activity ever, and outdoor biking proved to be equally damaging (lots of falls) and far scarier (going downhill on a road with cars).
Basically I want an outdoor elliptical machine that moves at a reasonable 5-8mph.
I have a (wobbly) pullup bar and a punching bag at home, though I'd need to install some bars in the yard to get an acceptable home setup.
Also, I was already using Google Drive since Docs/Sheets had already replaced MS Office. Don't need two services that do the same thing, especially when Dropbox has made no noticeable updates to its service in years.
Gym Memberships + Workout Classes - I used to be in a gym for 7-8 hours/week for years, but now I can't stand it. Realized I can get a comparably effective workout doing calisthenics at a local park while breathing fresh air and looking at the sunset and mountains.
Mobile Games - If I only have a limited amount of time on entertainment, I'm only going to play the best stuff out there.
Kindle/Ebooks - I discovered the library. Massive selection of free (well, tax-supported) physical books
Twitter - Used to check it 5x/day, but got no utility out of it aside from a few transient, impersonal messages to a handful of VCs, founders, and journalists. Deleted my account, never looked back.
Facebook - Socializing online using artifical interactions ("like", "share", one line comments) made me feel less connected with my friends/family. I still use their IM service from desktops, but refused to install messenger on my phone
Cable TV - Watching TV is incredibly frustrating with ads every 5 minutes.
Dedicated News Publishers - Used to subscribe to the NY Times, WSJ, Time, Atlantic Monthly, Scientific American, and the Economist. Although those publishers still have great content, I'd rather read what the internet surfaces (HN, Medium, random google searches)
Torrent Sites - There's finally enough free or affordable quality content to make torrenting less desirable, but maybe that's just because I've had an income.
Retail Stores - I now get all my food from Costco, Amazon (for misc spices), and farmers' markets for less. As a result, I buy fewer pre-made goods than I used to a few years ago and just cook in bulk. Far healthier and cheaper. I don't go shopping anywhere except to buy clothes every few years or medicine at a local pharmacy.
- Learn a relevant programming language (R, Python) + tools (ipython, anaconda, etc)
- Basic linear algebra (nothing more complex than multiplying matrices) and calculus (what are derivatives and integrals)
- Intro to statistics (just to know the vocabulary -- covariance, correlation, standard error, etc.)
- Rough overview of Machine Learning / AI as a whole and where its used in the world today
After that comes the second problem: "What do I do next?"
- coming up with several interesting but manageably small projects
- getting data for these projects
- access to quality advanced resources that can be consumed as needed while working on the project. A 3-month long MOOC on Neural Networks is an impractical resource. A well-written blog post (with code) or youtube video is far better.
MOOCs + textbooks seem to do better for the first problem if you can sift through all the noise, but fail at the second.
The biggest difficulty I have with learning data science is not how the algorithm or tools work, but the problem setup. Where is the data? How do I clean it? What insights can I draw from this? Which algorithms to use? What can I do with the algorithm assuming it works?
Most MOOC projects decide all this for you by giving you a set of tasks to do in order and skeleton code to work off of. Your job is simply to implement a small part of whatever algorithm you learned that week and press run. This way lacks creative development, exploration, trial and error, and critical thinking skills necessary when you go out in the real world.
Also, I think there should be more emphasis on publishing, even if your attempts are inaccurate. Push out a jupyter notebook to github of how you tested out a rudimentary monte carlo simulation on stock data. Or write a blog post with your attempt at determining how much silicon valley home prices will drop if 10K more family units magically existed in SF. Or try to code a random forest algorithm from scratch in a language of your choice. You don't have to be right, but publishing forces you to at least take a critical look at your work and think about the material deeply. MOOCs, at least from my experience, just encourage you to move on to the next topic the moment your code works, without diving too deeply.
Recently I've been thinking about setting up my on solar + wind energy generating systems in my yard just for fun. Mostly off the shelf parts that I just need to assemble together.
Any good hackspaces you know of that's accepting for complete beginners? I'm in the SF/Bay Area
But over the years I've seen the hills dotted with houses, untouched swamp land (great for the local wildlife) turned into generic commercial offices, and roads developed deeper into the outskirts. Noise and air pollution is getting worse. Parking and congestion is a nightmare anywhere there are new offices or apartments in old neighborhoods (SF/PA/MV). Public transportation takes hours to get you anywhere (though I'm optimistic about Bart expansions).
I haven't heard anyone in favor of housing development talk about these problems, problems that are bound to get worse if more units are built unless equally aggressive solutions are implemented.
Like others here have said, the Bay Area pressures career-oriented individuals to move here. Although I love the diversity of backgrounds, it is clear that the vast majority of my friends who moved here haven't developed the affinity or interest in non-career-related aspects of the culture/geography. They are not emotionally invested in everything else the bay area has to offer (I theorize that this is because their families and their childhood memories are elsewhere).
I'm not saying new housing shouldn't be built (affording housing for lower-income families/individuals need to be built ASAP). I can't afford rent here either and commit social suicide by choosing to live with my parents despite being in my mid 20s (fortunate enough to have this option). But I can't in good conscience support development in the Bay Area until it comes bundled with discussions on environmental protection, transportation expansion, tax changes (who is going to pay for the extra firefighters/police offers needed if population increases), infrastructure projects (pumping water into the bay, electricity, etc). As of right now, it is so singly-focused on lowering rent that these other issues don't seem to matter in comparison.
I appreciate sama's attempt to get us to call our representatives. As an active voter and lifelong bay area resident, I will do so too.
I think it was written in 2014, so it doesn't cover the latest trends in fracking or solar. I didn't know anything about the climate industry coming from a silicon valley background, so this was an ingestible yet detailed primer.
Not too many podcasts release revenue information publicly, so its hard to find.
There are two reasons I dislike donating:
1) publishers need to "beg" for money in every episode.
2) listeners are forced to have a charitable mindset and consciously/subconsciously assess if the publisher needs money (ie: no ads --> publisher can't feed kids --> donate)
DISCOVERABILITY:
- The best way to discover new podcasts seem to be through curated lists, word of mouth, or recommendations from other podcasts. Haven’t come across a recommendation algorithm that surfaces little known podcasts.
- iTunes, probably the largest distributor of podcasts, has strange algorithms. For example, “New and Noteworthy” is based on # of new subscribers, not # of downloads or reviews [1]. Android users don’t have iTunes.
- A particular publisher’s content is usually organized episodically instead of by topic. This makes it harder to dive deeper or learn more about something that piqued your interest.
REVIEWS / RATINGS:
- incentives to write a review is very low. Lack of reviews contribute to discovery problem Reviews are scattered across dozens of platforms, few of which have mass adoption
EPISODES:
- Most episodes are long -- 20 min to 2 hours. This makes previewing, skipping, and embedding into other forms of media harder.
SEARCH:
- Can’t easily search through an episode -- this is a general problem with audio searching
- Are podcasts even being indexed by any search engines if there is no transcript on the website?
PLAYER FUNCTIONALITY:
- Sharing an episode on FB/Twitter is hard -- users often don’t even need to sign in to use the podcast app. Sharing a segment of an episode is nearly impossible
- The “Skip” button is contextless and usually just jumps ahead a predefined number of seconds. It should be “skip ad” or “skip to next topic” While driving, operating the player means gambling with your life.
ANALYTICS:
- Primary metric is # of downloads, not # of listens.
- Ability to know when listeners drop off is nearly impossible
- Whatever little data is out there is scattered across every distributor.
ADS/MONETIZATION:
- Monthly subscription model means losing 95%+ of your listener base. Often this is still economical.
- Ads are doing well-ish ($10-$50 CPM), but fail to attract many advertisers because there is no tracking (unless there is an accompanied coupon code). CPM might be at its peak
- Too much free content available for a “netflix of podcasts” to work
PRODUCTION:
-High quality content, esp storytelling, is expensive or time-consuming to produce
ENGAGEMENT
- Users can’t do surveys, click on links, add tags, “like” or “share”, comment, rate/review, etc through most apps/players.
- No easy direct line to engage with publisher. Nobody goes to the website and writes a comment
- No easy way to engage with the community of listeners
- AMAs and live broadcasts are seen by a small % of fans
- Engaging with the content while passively listening (ie: driving, at work, at the gym, etc) is hard
There was a period of about a decade in which I had to wear custom-molded graphite orthotics (~$300) during most of my waking hours because my feet were supposedly "flat". Two podiatrists said that I'd need to wear them the rest of my life or get surgery, which would leave me in crutches for 6 months per foot (they won't do both at the same time).
I lost my orthotics in a flood, and unwilling to fork over another $300, I focused on fixing my feet. I started stretching my soul, walking around the house barefoot, and going for short (~100m) barefoot jogs on a grassy field. After a few months, I started wearing shoes without arch supports to force my feet to adapt. Managed to get rid of my foot pain within 2 months :).
Still struggling with the knee issue though. I've been treating running as an isolation workout on the knee joint -- my body can handle it just fine, but like in weightlifting, I can't isolate the same muscles/joints day after day. Sprinting any amount hurts, and I have to avoid all hills (going up is fine, the descent is killer).
While I figure out a way around the knee pain, I've recently been doing more metabolic conditioning (think Insanity or P90-X, but with fewer knee-destroying jumps). I experience a lot more fat loss this way, plus the workouts seem more holistic than running in the sense that my upper body is not neglected.
I think the best solution is to find one or two people who know their stuff to curate everything. Add community engagement mechanisms incrementally. For example, enable social media sharing and commenting before account creation, upvoting, and article submissions.
Wonder how a Quora/Medium model (invite-only) or The Information (paywall) would work for newsletters, since content isn't original?
I found some of the human-made playlists on 8tracks.com to be better than a lot of the algorithmically generated ones on Pandora/Spotify/Youtube, but music is highly subjective.
Been meaning to give product hunts collections a solid attempt: https://www.producthunt.com/collections
For book recommendations, I find an intellectually-minded friend on Goodreads who has similar tastes to me, and just go down his/her list.
I think a general pattern here is: there are positive deviants in society whose tastes and interests correlate with my own far stronger than the average's.
I used to sign up for an HN newsletter that'd only show the top posts. I thought that was an effective way of pushing only the best content to me every few days. Realized later that I enjoy reading a some articles that didn't get a lot of upvotes.
I also tried to create an open google doc where a small group of friends could share links to articles they found particularly noteworthy. Participation was 0, so I gave up and now just store interesting articles in my bookmarks folder. Messy.
"Emperor of All Maladies" - Siddhartha Mukherjee. An excellently written history of cancer.
"Endurance: Shackleton's Incredible Voyage" - Lansing Alfred. A true story of one of the last great explorations man has taken
2) Thanks for the link! Definitely looking forward to your blog post/postmortem (don't let BetterExplained die!). Have you found learning to be less enjoyable by writing about it, or do you end up discovering a dozen other tidbits of math magic you want to share with everyone?
Two more questions if you don't mind :)
3) What are your thoughts on word problems?
For example, in most linear algebra textbooks, you are given matrix and are asked you to process it. Rarely are you given a word problem and are asked to think through the entire process (data and operations matrices setup --> processing --> meaningful end result).
4) For inspiration, what are you experiences explaining concepts in a cross-disciplinary manner?
When I was a student, I never understood why a concept is important. Homework problems were abstracted out of all real-world context to train for mechanical problem solving. Only now, after exploring data and writing algorithms in health, journalism and finance, have I finally been able to answer the question I always had as a kid: "why is this stuff useful?"
Two Questions:
1) What are your thoughts on interactive content like ExplainedVisually? I've been thinking about doing something similar for data structures / algorithm topics. How much of learning math concepts is exploratory vs learn-by-doing?
2) Are you running BetterExplained as a side business, or full-time? And if you're willing to share techniques and numbers for the entrepreneurial HN community, what are some things you've done to market it, monetize, etc and what were the results?
Thanks!
Computers:
- Navigating the file system is confusing, esp on a PC with a lot of default folders that don't need to be touched
- Drag + Drop relocates local files, but copies if you're uploading to the web/external device. Same action but different results --> requires contextual awareness
- Setting up a new device with the essential software
- Debugging printer and wifi problems.
Web:
- Password recovery/management
- Figuring out google drive
- the concept of shared files vs attaching everything in emails
- using anything aside from google search, youtube, and gmail
- using product reviews to identify the best product is not how they shop in real life.
Smartphones:
- discovering your own number
- making text larger
- managing notifications / changing settings on an app to app basis
- uber/lyft. Large distrust of strangers
I used to buy a few kindle books (or pirate the ones I couldn't find digital copies of), but I found out I read both more and faster with tangible books. And the three-week rental period (with the option to renew for up to six more weeks) encourages me to actually finish books in a timely manner.
My library also hosts cultural shows and has a small museum -- currently there are several pre-modern human skulls on display.
Plus, internet speeds hit 50Mbps+. Never need to go to a local coffeehouse again.
The only odd thing about my library is how ebook rentals (3-week digital access) can be out of stock. Don't know what kind of licensing deal they have with publishers, but this seems pointless.
On a different note, what I've noticed is how easy platforms like this help fight piracy:
Netflix + HBO --> fewer torrented movies/shows.
Steam/Humble Bundle/etc --> fewer illegal games.
Libraries --> fewer pirated books.
Youtube/Spotify/Pandora --> no longer need to maintain a library of illegally procured music.
With the right combination of price, selection, accessibility, and convenience, piracy to me has seemed less appealing over the years.
I usually perform better on take home interviews, but 90% the time I'm unwilling to dedicate what are usually days for just a chance to be accepted at a company I may not even want to join. I think many employers use the take home interview as a screener without realizing they need to first cultivate in the candidate the enthusiasm and willingness to complete it.
Some ways companies can encourage me to actually complete the take home interview:
- Provide compelling information about the job (estimated pay/equity, meeting the team, evaluating culture-fit, getting me excited about the problem, etc). Lots of companies save this step for the courting process that happens after the technical screen.
- Pay me a modest amount just to complete it, regardless if I pass or not. I don't care about the money, but at least it won't feel like wasted time.
- Make it way shorter, but then it might be useless.
--
What I'd prefer instead of all this interviewing is still:
1) meet the team, evaluate culture fit, etc.
2) discuss past projects in detail, maybe do a code evaluation if relevant
3) contract for the company for 2-4 weeks at market rate pay. Hell, make a 10-week "internship" out of it, whatever.
4) receive offer (or not)
This is not always practical or scalable, but I've gotten offers 100% of the time this way, including at YC companies and other startups. And from the company's perspective, I think it extricates any unreasonable expectations they have from prospective employees. It's always tempting to look for unicorns when you have hundreds of resumes to choose from, but when you have a likable contractor doing good work, there's no reason not to give him/her an offer. Plus, the employer can evaluate the single most relevant skill in an employee -- the ability to learn.