RubenvanE

Done. :)

I will start looking into how to make the code open source before launching the project officially!

Unfortunately I won't be able to submit the project to https://hnpwa.com since they only allow (more or less) direct copies of HackerNews to their directory. [1]

[1] https://github.com/tastejs/hacker-news-pwas

This is an interesting idea! The problem is that you regularly get 8+ levels deep, so that's a lot of clicks. Maybe two buttons (for parent and for top-level parent) might do the trick?

Yes, need to fix this. Thank you!

Yes, I am aware of the issue (see my recent comment on one of the parent comments).

I am afraid that it is an issue with the way some browsers render the CSS sticky position. If you have any CSS knowledge, I would greatly appreciate your input!

For my information, which browser do you use?

Thank you, this truly warms my heart!

I have poured too many hours of my life into this little detail. It still doesn't work great though:

1. It is hard to click on on mobile devices

2. It sometimes renders behind the comment or is not clickable (not always, just sometimes). I am afraid this has something to do with the way browsers render the CSS sticky position, so I might not be able to fix this. I am not very experienced in CSS though, so if you've any advice on how to fix this I would truly appreciate it!

Because of the above two issues I might need to go back to just a fixed minus button. :(

Thank you, I'll probably post it on the main page when It's a bit more finished.

It depends on the webpage, if the webpage is not an article it just redirects to the webpage. If there is a book icon before the the title the webpage has been converted to an article.

That's weird! I've used it mainly on Chrome so far and I've not yet encountered this problem. Does it keep happening or was it once?

HackerNews has an open API (https://github.com/HackerNews/API). It doesn't say anything specific about the legality and there have been many HackerNews clones in the past. So as far as I know it should not be a problem, but if anyone can correct me I would love to hear so!

Thank you! I mainly build it to learn GraphQL and Postgres for myself. But if people are interested, I could share the source in the future.

I had the exact same problem. There are a lot of interesting discussions, but its way easier to distinguish the interesting from the useless discussions when they're done instead of ongoing. I find that I read more carefully and focused when the discussion are already crystallized, instead of just browsing the latest discussions for some dopamine.

It takes all posts that were posted on a certain day (based on your local timezone) and shows these posts ordered by their score. To keep it clean only posts with a score of 50 or higher are shown. (I might change this number though, if people would like to see less popular posts)

E.g. the main page is always 'yesterday', so for me it shows the posts that were posted between 00:00 and 23:59 on Wednesday, but if its already Friday where you are it will show the top posts of Thursday.

These sentiments are quite familiar to me. That's why I've been building my own HN copy (https://hackerdaily.io). It shows all posts per day, which has a few benefits:

1. It doesn't become an addictive slot machine that updates every minute.

2. The conversations have already happened, which makes it easier to see which threads are interesting and informative and which are not.

It also uses an outline.com like style for the articles, which removes the tracking and spamming and places the focus on the articles itself.

It's still in active development and so far I've only used it myself, so all feedback is welcome!

Ps. It’s an PWA, so you can add it to your homescreen and use it like an app.

The foundation package sounds a lot like the Lion components that ING Bank open sourced last year. It is indeed super handy to not have to worry about all the browser compatibility and accessibility issues!

But is spamming not in its essence 'unethical'? It might not be illegal, but it is morally at least questionable and many will say it's unambiguously immoral.

SEO ‘hacking’ is clearly positive for the businesses doing it, but the benefits on a larger scale are less clear cut.

Do you believe strategies such as outlined in the article are a net positive, negative, or a neutral tactic for search engines and the web at large?

And those are just the high-tech, high-profile projects.

Even a mundane bridge near a small town saves thousands of people a day of making a detour. This might not sound 'impressive', but imagine a bridge that gets used by just a thousand people a day and that replaces a detour that takes 15 minutes longer. That's a 1000 people * 0.25 hours * 2 times a day = 500 human hours saved per day, or over 180.000 hours saved per year. With a lifespan of 50 years equals that to close to 10 million(!) hours or 4500 'working years' saved. This is an example of just a small bridge near a small town, just imagine the benefit of larger bridges (and other infrastructure projects) near the big cities.

But to what extent is it 'publicly distributing' when the app has only 200 users. It sounds more like an extended beta than a 'clever' way to circumvent the app review process.

And, as the author extensively explains, the major sin of Apple is not that they removed the app from Testflight, but that threaten to suspend his developer account completely and give no explanation _why_, even after multiple attempts of the author to reach out to Apple to solve this situation.

Wow, the service stops on May 28th, less than a month away.

Just out of curiosity, have there been more (relatively successful) companies with a paid subscription plan that announced that they would, within a couple of weeks, shut their services down? As an outsider it seems quite drastic to me, given that the daily costs of keeping the servers running a bit longer is negligible.

You're right, but technically tu quogue is a type of ad hominem attack. So OP used an ad hominem tu quoque fallacy to discredit the author.

Even though this might not be the strongest argument from a purely logic point of view, it can still be a valid critique on the author.

I got the same 403 message.

Luckily outline.com was able to fetch the article: https://outline.com/yecexc

While I agree that the fee might be high, the comparison between the two is is a false equivalency.[1] A transaction fee is something completely different from an interest fee. The first is a one time fee while the other is continues over time.

[1] https://en.wikipedia.org/wiki/False_equivalence

Your analogy with the energy market is quite good, the only thing I would address is that coal plants aren’t more energy efficient than gas plants (roughly 40% and 50%, respectively). The reason coal plants provide the base load is that the raw material, coal, is cheaper than gas. However, with falling oil and gas prices this might change in the ‘near’ future.

The reason the US military (among others) buys so much equipment even during relatively peaceful periods is to keep alive the ability to produce massive amounts of tanks etc. whenever they really _do_ need them. So just not outsourcing these kind of essential goods to foreign nations might be already a great leap in the right direction.

A better way to find the 2020 equivalent of $5.00 would be to look at the change in Consumer Price Index (CPI) over the past century.

The CPI changed from 9.9 in 1913 to 255.7 in 2019. So $5.00 dollars in 1913 is equal to about $130.00 in 2020.

Awesome product! I was quite surprised at how good it summarized Dutch articles as well.

Are you planning to launch an API anytime soon?

I think you meant to reply to this post https://news.ycombinator.com/item?id=22166846. :)