HNHacker News
TopNewBestAskShowJobs

poletopole

198 karma · joined October 14, 2016

modum patio tace memento
submissionscomments
poletopole··on Ask HN: What are you planning to unlearn in 2021?
I once got fired for printing out and posting a jackie chan meme about drupal on my cubicle. Everyone thought it was funny as shit except for my boss.
poletopole··on Ask HN: What are your goals for 2021?
Survive.
poletopole··on Ask HN: Why don't companies want to employ people on 4 day per week contracts?
I worked part-time (32 hours a week) for 7 years. What I discovered is that most larger companies simply don’t do this schedule because the larger a company is the higher the competition. That’s why I only use CL for applying to jobs because I have more leverage to negotiate my hours.

Also, even though you may be able to start part-time, as the company grows, you will become a more integral employee, and your responsibilities will grow which materially implies you’ll be expected to go full time eventually. The best way to get around this is to work 5 days a week but fewer hours each day.

poletopole··on Show HN: Every day a piece of Classic Art in your email
Sorry for the delay. I actually appreciate most art but pragmatically I’m most interested in art that software incorporates into its UX. Typically this is usually vector art or digital art, but a good example is what you might find on Dribbble but honestly most stuff on Dribbble is mediocre and derivative. So what I’m saying is that I’m interested in seminal art curations.

I signed up btw ;)

poletopole··on Show HN: For coders to connect and share knowledge
I miss HN office hours and I was contemplating how to make a better version of recent, but this could be a nascent form of it.

Can you tell us a little more about this service or your impetus or inspiration for creating it was? It reminds me of microblogging version of Forest. One thing I might add is, be sure to engineer a way to keep the diversity and population size of the population stratified into small samples, otherwise new users will gentrify the service.

poletopole··on Show HN: Every day a piece of Classic Art in your email
This is brilliant in my opinion. But I wouldn’t limit it to just classic art, but it’s a good place to start!
poletopole··on Ask HN: Is there a note taking app you're 100% happy with?
I’ve been using Ulysses for about two years and I’m addicted now. However, it’s not the the whole solution I’m looking for; ideally I would like a Muse + Ulysses + Gladys app, so I probably will just make it myself this year.
poletopole··on Ask HN: How Stable Is Pijul?
Thank you for taking the time to answer my questions. I should have asked also if Pijul’s current API or user facing semantics will change with 1.0?

I admire how challenging developing Pijul must be, especially when some of us here critique it without bothering to understand it, so my hat is off to you! But I think most of us here are eager to use Pijul, myself including.

poletopole··on Ask HN: What are your best purchases in 2020?
Awesome, thanks!
poletopole··on Ask HN: What are your best purchases in 2020?
The new iPad Pro and magic keyboard. Look into Tailscale, it makes it incredibly easy for me to ssh into my mac or a VPS. As for apps for iPad, I highly recommend the following:

1. Muse: awesome workspace app.

2. Ulysses: best MD app hands down.

3. Gladys: good clipboard.

4. Kodex: best ipad code editor.

5. Prompt: best ipad ssh client.

6. Working copy: good git client.

7. Kaleidoscope: diffing files.

8. Jira: pretty well done.

9. Documents: good for ftp.

10. Copied: good for bookmarking.

11. Goodnotes: best notebook app.

I’m seriously thinking about selling my imac because I haven’t used it since I got my ipad pro. I think I’ll sell it and build a RaspPi machine, ha.

poletopole··on Ask HN: What are your best purchases in 2020?
I’ve been looking for a good drawing app for ipad. Is ProCreate easy to learn?
poletopole··on Ask HN: Book recommendations to learn functional programming
The only book I found helpful in learning functional programming was Lambda calculus and combinators by Hindley. Although you can find it online for free it’s worth buying. What I discovered was that books on Haskell simply weren’t enough to understand the language, you really do need to understand lambda calculus first.
poletopole··on Ask HN: How to develop empathy?
It’s okay to be indifferent, perhaps it’s natural for humans to be ruthless. If your friend is so self-absorbed as not to notice he is about to step into poop, then he will step in poop, just not while you’re around to smell it.

Buddhists are the most indifferent yet compassionate people I know because compassion requires impeccable mental silence at all times, and after so many years you realize it really doesn’t matter if your friend knows he’s about to step in poop or not. All that matters is that he learns his lesson, that he was too important to not notice the poop himself, and that’s not something you can tell him, he has to learn that lesson himself.

poletopole··on The Graph has just launched its decentralized network
As someone whom has been working on a distributed and decentralized HTTP multiplexed “L8” protocol for the past several years, I’ve seen many P2P protocols come and go, but none of them actually gain traction for long. In my opinion, the problem these protocols run into is that, the web needs protocol diversity, not “one true protocol”.

Once we have a culture of protocol diversity, where protocols are built and compiled in a day rather than years, then and only then will we see things like a truly semantic web that isn’t owned by Google.

It feels like these new P2P protocols aren’t designed with distributed applications in mind. This is because we need an entirely new type of symbolic/logic/linguistic programming language and runtime that augments natural language rather than boil the ocean with ML to understand it and represent its semantics topologically, not stochastically. Ideally, this language ought to model each user’s I-Language as a universal grammar that is isomorphic to all other dialects, meaning one user’s grammar can render to any other grammar. This is easier said than done, but not impossible. Here, a dialect defines a protocol within a network.

I always find it funny when I login into Mastodon and watch the federated public timeline, it’s all toots written in foreign languages I don’t understand. Only once we as an industry start augmenting and modeling natural language in a way that isn’t ancane, can a distributed/decentralized protocol suite even have value for the world. What’s the point if half the world doesn’t understand the other half? Unicode is garbage and we can do better, much better. As a developer, I would say about 30% of my time debugging was related to character encoding problems and about %75 of my overall time was debugging, not developing.

Being distributed and decentralized isn’t enough, we need a toolkit that simplifies development to the point that a developer can release a new application in a week, not months or years, all 100% bug free. The way forward is inverting how we network applications, such that we only write pure business logic on the network datum level on a per entity basis and write protocols with those entities (entities whose schemas need not be public) with zero code, meaning only the network topology of an application (events and protocols but not states) instance is needed explicitly to be public, which can be compiled into a repo and distributed, and would generally be adaptive in ways I won’t elaborate here.

However, I would be hypocritical if I did not say that Graph is at least in the right direction and at least uses other protocols like IPFS and doesn’t completely reinvent everything. What I’m saying is that our woes as both users and developers runs much deeper to the point that it’s almost a conspiracy. The problem is that introducing a new protocol in today’s climate is almost seditious because, unlike what I’m workin on, most if not all P2P protocols make claims that they are revolutions (not reformations), so the day when we can have a distributed and decentralized semantic web that can serve requests over HTTP/1.1 easily, everything else is just not pragmatic nor rational.

poletopole··on Ash HN: Work After Tech
I used to go to the subway down the street from my house. There was this old strange employee that at first gave me the creeps, but eventually we became casual friends. On his last day before retirement he gave me the book he wrote. It turns out this guy was a millionaire and had a very diverse set of skills.

His book said that even though he was an accountant, stock broker, song writer, etc he preferred to work in the restaurant business because he liked working with the public and being social.

What I learned from him is that you really aren’t your job. Even if you have to work at subway to enjoy life, don’t think less of yourself ;)

poletopole··on Ask HN: Do You Believe in Aliens?
I believe in much stranger things that are inclusive to aliens.
poletopole··on Ask HN: Why is naming things in programming so hard?
Programming is very symbolic whereas natural language is more phonetical, so using one in terms of the other will always be awkward. I feel the best strategy is just to let variable names, for example, be pithy or terse and just use comments to compensate; this is because I’ve found comments are still necessary even with long semantic variable names.
poletopole··on Ask HN: Have you tried leaving social media
So this turned into more of a rant so I apologize in advance. I left facebook and social media (excluding HN of course) in general 10 years ago. The sad reality is I lost every friend I had to that hell hole. The straw that broke the camel’s back was the day facebook notified me that a Serbian had accessed my account, so I scrubbed all my content, unfriended everyone, and deactivated the account except for one post warning not to trust any future posts.

I’m even considering leaving HN after taking a two week break from it. The problem with all online forums in general is that they sour once their population reaches the onset of chaos. This phenomenon is not as subjective as it might sound; it’s loosely related to entropy in information theory introduced by Claude Shannon, but that’s a conjecture beyond the scope of this thread. Humans historically work at their best together in small groups or tribes for the same reason why smaller classroom sizes produce better grades, meaning that our capacity for communication increases with our level of autonomy.

I’m frankly surprised why that fact isn’t salient among social media services today, but I’m sure there’s a good reason. I know Mastodon is taking distributed social networking seriously, but honestly Mastodon is very awkward, it’s kind of like being in a small waiting room alone with another person where you can only communicate in terms of orthogonal and pithy announcements, rather than form a natural conversation.

I’ve been thinking a lot about the elements that would make for a natural, healthy, and conductive social networking service quite recently, but my largest question is how much of a factor is the underlying platform technology itself a measure of success in engineering a social network?

I was surprised when my nephew told me not too long ago that his only friends are ones he meets in his video games; having grown up in the early 90’s this is bizarre to me, but also is provocative in the sense that it reminds me of how random sampling works. An individual may more likely to form a significant lasting social bond as part of a random sampling of a population of peers through a controlled trial process, such as a video game, than they would had the selection process been through biased selection events such as systematic population nepotism, proximity, or stratification. For example, I hate my neighbors, I don’t miss my facebook friends, coworkers and couples are a drag, and I could honestly do without relatives. I think that finding friends as an adult is hard for many logical or practical reasons, but may just be because in life we our forced to abandon our sports and adopt absurd “hobbies” instead. Having said this, I can take pity on my neighbors rather than hate them, because now I realize that sadly the only socially acceptable way to play as an adult these days is with annoying pets.

poletopole··on Ask HN: How to efficiently import Excel/CSV files
I’m not sure I understand the problem space here, but consider AWK. It really changed how I think about any problem involving tabular data parsing.
poletopole··on Ask HN: How do you find new music?
I have to second soma.fm. I almost forgot about it though, it’s been a decade at least, damn.
poletopole··on TikTok tried to manipulate me – uninstalled
I received a verification sms from Tiktok not too long ago. Some runt used my phone number to sign up for Tiktok. I don’t even know what Tiktok is.
poletopole··on Ask HN: What is your biggest problem?
Fly me to the moon and let me play among the stars
poletopole··on Ask HN: Has anyone else just gotten bored of the Internet?
Yes, Carlos.
poletopole··on Ask HN: Has anyone else just gotten bored of the Internet?
I read Ted Kaczynski’s manifesto a few days ago out of curiosity to find answers about why serial killers don’t make the media anymore, not even on fbi.gov. I asked my mother the same question since she routinely calls the FBI to interrogate them for reasons I won’t discuss here. Even she was surprised by the question and thought it very odd as well. It seems that the “new speak” for serial killers changed after the patriot act. They are now labeled as “terrorists”.

I found Kaczynski’s manifesto actually is harder to find than one would imagine. There are many falsified versions of it circulating online. When I did find the original, it became abundantly clear why talking heads frame him only in context of rumors and never about the philosophy behind the manifesto. What he wrote is a hard pill to swallow, all for the same reasons Castaneda’s works are all but scrubbed from mainstream culture. For even a mature adult, for example, a software developer, to concede that their entire industry in practice and as a science is a manufactured and dogmatic “surrogate” ritual, is the reason they are depressed and humiliated, would undermine their purpose and significance in life.

The easy answer for the OP is that simply SEO ruined the internet. The harder answer might be that, in Kaczynski’s terminology, our “power process” as individuals is all but lost. Empirically, finding relevant resources online in the 90’s required considerably more effort than it does today. If I’m stuck fixing a bug today all I have to perform is a painless Google search. Why then would this be problematic? As Kaczynski argues, the more we remove the effort to perform our work, the more demoralizing it becomes. This is because, although the technology (Google) is useful, it nevertheless systematically removed the power and purpose derived from the challenge and art of that technicality. Software engineers have the distinction and privilege of being first witness to this complex deprive the rest of humanity of purpose because we’re the ones programming it. Or so the story goes...

So to answer the OP’s question: the internet is now boring because it’s no longer a challenge.

poletopole··on Ask HN: Why are ML developers leaving TensorFlow for Pytorch?
I stand corrected. Take Polymer for example. I’m very surprised, despite all the negative commentary and controversy surrounding web components, that Google has remained loyal to their Polymer project and even developed it further into LitElement. I think what happens is that Google will bring an OSS project to market and then after a few years the community takes over, so there’s an illusion of “abandonware”. But it’s hard to say whether Google just moved on to something else first, for example from GWT to Angular, and then the community took over or if it’s the other way around. So there’s this implication that Google does abandon their projects or at least stops funding them even when that’s not really the case, but the implication alone almost becomes a self-fulfilling prophecy. In fact, at work I suggested we even do a project with LitElement over React and the reason the team went with React was because they thought Google would abandon the project. ¯\_(ツ)_/¯
poletopole··on Ask HN: How do you find help?
I may have got lucky but I found a psychiatrist who was life changing with Zocdoc. I recommend finding a private practice psychiatrist first since they can refer you to a specialist for therapy—doing the reverse is much harder. My sister is a clinical psychologist in the Navy who works with many PTSD patients, so finding a therapist with military experience would make sense in your case; she’s a proponent of exposure therapy. What I discovered was that the most helpful doctors are always in the most affluent part of town, so keep that in mind, it’s worth the drive.
poletopole··on Ask HN: Which general programming language is good for learning/exploring math?
I’m learning Haskell right now and had the same question until I found Hindley’s Lambda-Calculus and Combinators: an Introduction book. Many here are recommending Julia but I’ve used Julia enough now to conclude that it isn’t as mature as Haskell and can be frustrating at times. Haskell, when it’s foundations and history are well understood not only challenges how you think about programming but also of mathematics.
poletopole··on Ask HN: When and why do you choose to use a desktop client vs. a web client?
When there is a true native and solid/stable desktop version of an app I usually end up using that over the web client. The reason for this is because as a web developer I know what can go wrong with web apps, so it’s partly an irrational fear I suppose.

About a year ago I decided to splurge and upgrade my development process with native apps like Ulysses, Git Tower, Kaleidoscope, CodeRunner, TablePlus, etc and discovered that native desktop apps can actually be well worth the money, so I never looked back since then. There are always exceptions to the rule: for example if 37signals had not closed Backpack down I would still be using it and Jira’s native client is crap compared to their web app version.

poletopole··on Ask HN: Why are ML developers leaving TensorFlow for Pytorch?
I asked my friend who is both a web developer and does ML if he preferred pytorch to tensorflow and he confirmed what I had read online, which was that pytorch is better for parallel ML and other use cases that TF isn’t. It doesn’t appear that all TF libraries are made equal. An of course, Google does have a long history of abandoning their open source projects. However, my friend did admit that TF excels better at low level matrix operations.
poletopole··on Ask HN: How do you cope with mental illness as a founder and/or employee?
Back in November 2013 I spontaneously developed what would become permanent depersonalization/derealization dissociative identity disorder. I have three weeks of missing time I can’t account for when it triggered; according to my family I was at home the whole time while it happened, as if I was sleep walking, but no one noticed.

When I came to, I found myself in a very literally dark, distored, and morbid hell-scape. I was very confused and although I still had memories of being “me” I no longer associated them with whom or what I was now and likewise for family and friends. I was a stranger to myself and everything and everyone I had ever known.

I went to see my primary and the nurse went to take my pulse and it was so high that she brought in the doctor to take it multiple times because they thought the machine was broken. I was recommended to a psychiatric clinic by my home and went there for about eight months before I switched to a private practice psychiatrist. This was because clinics are too afraid of being sued and thus their doctors rarely are willing to experiment or go off-label on their patients in order to cure them.

After almost 8 years of experimentation and trying pretty much every medication on the market and with the help of my doctor, I can say I have recovered to my satisfaction. My disorder is rare and thus isn’t well documented in the DSM and therefore is hard to treat and diagnose. I still, from time to time, have periods of “missing” weeks that I can’t account for but I’m still functional and appear normal during those periods.

I learned many lessons about the medical industry over the years, but the most important is: “Only experimentation cures patients”. You have literally 5 minutes per month with your psychiatrist because they “prescribe“ whereas psychologists “talk”. Buy the best healthcare you can afford and only go to pharmacies and doctors in very affluent neighborhoods. If a medication or class of medications, such as SSRIs, didn’t work the first time, they won’t work a second time. Be patient, because your doctor has other patients—this means you have to be responsible for doing your own homework about your disorder and what treatments might be effective for you. This also means you have to learn a considerable amount about neurochemistry and biology sadly and keep a journal of controlled experiments to determine what factors are significant or not. Don’t be afraid to fire your doctor if it’s obvious they don’t value your time, such as routinely being late for appointments, or ignore your every word and just follow a scripted dialogue with you. I wish I could tell the readers here all that I’ve learned, but for brevity’s sake I won’t here, but I can comment on any responses.

Lastly, my advice is do NOT tell your employers nor your coworkers about your condition. They won’t care nor understand what you’re going through and they will use it against you the first chance it becomes convenient for them. I’ve seen this happen time and time again with past employers; they love to say “Oh, he was crazy” as the reason they fired a developer who dared challenge their authority. This is the unfortunate stigma the mentally ill have to contend with on a daily basis and only makes things worse for those of us whom are struggling.

One final word: true mental illness is a life long struggle and what has kept me from giving up so far is remembering, “There’s always something better”. No matter what stage of recovery you’re at, you can always be a better you—never stop improving.

← PreviousPage 4 of 7Next →