151–160 of 276 posts
Earlier quoted context omitted.
Danny Crighton has a CS degree from Stanford. I'm not entirely convinced by the article either, but your criticism would be a bit more credible if you managed to avoid being factually wrong.
You can have a CS degree and never code.
Earlier quoted context omitted.
I assume you've given uMatrix a shot?
Nope. I was using uBlock origin on Firefox with dynamic blocking and quite a bit of the provided blacklists enabled. I just gave quick uMatrix a look, and it does not seem fundamentally different to me. I like that I can easily add some globs and use version control on the blocklist without having to import-export my settings. Qutebrowser is a very interesting browser. It's easily scriptable using a config file in py…
* facebook.com * block
* fonts.googleapis.com * allow
* fonts.googleapis.com cookie block
* fonts.googleapis.com frame allow
* gstatic.com cookie block
* gstatic.com frame allow
* gstatic.com media allow
* gstatic.com other allow
* gstatic.com script allow
* gstatic.com xhr allow
* twitter.com * block
This whole "RSS died" narrative illustrates a basic problem with how we understand the internet. Since there's no unified unbiased sort of "index", like GDP, to tell us just what proportion of users on the internet uses RSS anymore, we're stuck having to rely on socially propogated ideas about the state. This leads to all the typical problems with socially defined truth, such as drift as a consequence of incentives (…
Surely Google Reader had numbers that influenced the decision to shut it down?
Earlier quoted context omitted.
Danny Crighton has a CS degree from Stanford. I'm not entirely convinced by the article either, but your criticism would be a bit more credible if you managed to avoid being factually wrong.
You can have a CS degree and never code.
Earlier quoted context omitted.
For machine learning there is the Arxiv Sanity Preserver, which can recommend new papers based on your library. http://www.arxiv-sanity.com/ It didn't really preserve my sanity, though. I'm reading those papers in my spare time, and it got me to read many more papers than I ever did before.
Not seen that before (my field isn't comp sci or ML), but it looks great! Obviously web interface is the way to go for a serious project, but I actually found it fun learning how to implement an interactive command line app. It's incredibly responsive and has a four-key interface, so if you are good at skim reading you can train it with a lot of papers in very little time.
No one has mentioned NewsBlur yet. It's fantastic as an RSS reader, and provides intelligence training (filtering) for only subjects you might be interested in, allowing you to tame high-rate RSS feeds quite easily: http://newsblur.com/
I still don't quite understand the fascination with video content: It's completely synchronous; a 10 minute video takes up 10 min of my time, with limited ability to consume more efficiently. Textual information puts the control back in my hands, with the ability to aggregate content and quickly consume. RSS has been the best way to do so while avoiding the dross that every social network has devolved into.
Earlier quoted context omitted.
Surely Google Reader had numbers that influenced the decision to shut it down?
I see Google Reader mentioned a lot when virtually anything RSS-related comes up. I missed out on the whole RSS craze (prison will do that to you). But I have always wondered -- so many people lament the fact that Google killed it -- why hasn't anyone forked/cloned it's UI and feature-set? You don't want to just straight up copy it and wake the Google lawyers from their nests obviously, but why not make something tha…
There are clients that kind of did, but one of the cool things about Google Reader that a lot of people miss is the social feed. You could share articles you found interesting and your shares would show up on the feeds of anyone who followed you as just another RSS source. It's just an extra folder of articles from your GChat friends that you can read at your leisure.
And it's people selecting articles from their normal reading rotation, so it was refreshingly clear of clickbait. People usually shared stuff because they thought their friends would find it interesting rather than distributing content designed to "go viral."
Once the ecosystem fragmented nobody has the critical mass to replicate the social features. It was nicer than what Twitter or Facebook do because it wasn't in your face.
No one has mentioned NewsBlur yet. It's fantastic as an RSS reader, and provides intelligence training (filtering) for only subjects you might be interested in, allowing you to tame high-rate RSS feeds quite easily: http://newsblur.com/
Can this be self-hosted?