4 ms·

Show HN: HNText – Top stories daily cleaned for readability

36 points by bckmn 12y ago

m90 12y ago
Interesting idea, http://www.hntext.com/#543c2fd8ba03530200000008 http://www.hntext.com/#543c2fd8ba03530200000008 has "broken" special characters though.
- Meekro 12y ago
  I'm having this problem as well. I took a screenshot just in case op sees it normally and doesn't know what we're talking about: http://i.imgur.com/0GlHrlc.png http://i.imgur.com/0GlHrlc.png Aside from that, looks like an interesting service, though!
JacobJans 12y ago
This is absolutely horrible and shows complete lack of respect for copyright. It should be taken down immediately.
- Houshalter 12y ago
  This isn't qualitatively different than readability or numerous other extensions. The only difference is he is doing it server side.
  - declan 12y ago
    > The only difference is he is doing it server side Ah, but from a legal perspective, this makes all the difference. It is, I'm sorry to say, copyright infringement and goes beyond what fair use permits.
    - petercooper 12y ago
      You're right, but I wonder where the line is drawn with things like online RSS readers. They're basically doing the same thing, but just behind a username/password.
      - declan 12y ago
        >line is drawn with things like online RSS readers For starters, I'd look at the Supreme Court's opinion in Grokster. Here it is in HN-suitable pseudocode. :) IF $distributing_product_used_by_others_to_infringe == FALSE THEN FIND_NO_LIABILITY ( "None found" ) ELSEIF $distributing_with_object_of_promoting_infringement == TRUE THEN FIND_LIABILITY ( "You're inducing infringement ") ELSE IF $actual_knowledge_of_infringements == TRUE and $failed_to_block == TRUE THEN FIND_LIABILITY ( "You're a contributory infringer ") ELSE IF $capable_of_substantial_noninfringing_uses == TRUE THEN FIND_NO_LIABILITY ( "None found" ) ELSE FIND_LIABILITY ( "You're a contributory infringer ") END END END How this applies to online RSS readers is left as an exercise for the reader... UPDATE: A copyright lawyer friend points out via email that I neglected to include vicarious liability. I'll leave the above pseudocode as-is and note vicarious liability here as a bugfix/todo.
    - Houshalter 12y ago
      Yes I understand that, I said it isn't qualitatively different. If the web page somehow distributed javascript that would do the same thing client side, the website would appear exactly the same, yet would be perfectly legal. The comment I was replying to wasn't arguing that it was illegal, but actually wrong and "lacking respect". I was just pointing out that he should have the same opinion towards things like readability. Also I think readability did do it server side, and so do other things like RSS readers.
      - declan 12y ago
        Without taking a position on Readability, there's more to the calculation than client side vs. server side. For more see my post here: https://news.ycombinator.com/item?id=8455648 https://news.ycombinator.com/item?id=8455648
jawns 12y ago
Three criticisms, in descending order of importance: 1) The length of these excerpts goes a bit beyond what is typically considered fair use. You're pretty much reprinting the entire articles. So you may run into copyright infringement claims. 2) It looks like you're scraping the content, and your scraper isn't up to the task. One of the excerpts reads: "Your email has been sent. An error has occured and your email has not been sent. Please try again. • You can't enter more than 20 emails. • You must enter the verification code below to send. • Invalid entry: Please type the verification code again." 3) Borrowing the multicolumn text layout predominant in print is great when the columns all fit on the screen, but I found myself having to constantly scroll up-down-up-down-up-down to read these. So, in sum, you're only presenting the top stories successfully some of the time; when you are successful, you're infringing on copyrights; and the layout is not actually cleaned for readability.
btbuildem 12y ago
Speaking of readability, have you heard of https://www.readability.com/ https://www.readability.com/ ? Works wonders on all but the most stubborn webpages.
- eli 12y ago
  http://www.diffbot.com/ http://www.diffbot.com/ offers this service as an API
JustThrowinAway 12y ago
My first thought was how these articles were in no way more readable than the originals. Does anyone prefer a multi-column layout?
- singlow 12y ago
  Multicolumn is fine as long as the vertical height is less than the browser window. For long text this is really hard to make work in the browser unless you take over the pagination with javascript. This is not a good example of using multicolumn text.
andremendes 12y ago
http://getpocket.com http://getpocket.com do this "cleanse for readability" to the saved links but I think they got closer to readable than you did. If you get to avoid copyright infringement, and get rid of this multi-column layout, I'm sure it would be better.
dotsam 12y ago
The generated podcast is also presumably a violation of copyright (i.e. it creates a derivative work).
Xeoncross 12y ago
Neat idea, I'm sure you are violating fair use - but it is a neat little mashup of scrapers and text-to-speech. I won't ever use it though. I have a bookmark for cleaning up pages.
shmapf 12y ago
For those wondering about the usefulness of this. It should be great for those who commute on the tube or somewhere else where there is no phone signal. This lets you load a single lightweight page before boarding, and then be able to browse hn easily. Would be ideal if it was sent out in an email though so you don't have to remember to load the page every time.
lziz 12y ago
Cleaned for readability, eh? - http://i.imgur.com/faqQrQi.png http://i.imgur.com/faqQrQi.png
kolev 12y ago
I think a script that does this, but puts in your Pocket with an HN label would be even better - you keep track of what's read and what's not as a bonus.

36 points by bckmn 12y ago

m90 12y ago

Interesting idea, http://www.hntext.com/#543c2fd8ba03530200000008 http://www.hntext.com/#543c2fd8ba03530200000008 has "broken" special characters though.

Meekro 12y ago
I'm having this problem as well. I took a screenshot just in case op sees it normally and doesn't know what we're talking about: http://i.imgur.com/0GlHrlc.png http://i.imgur.com/0GlHrlc.png Aside from that, looks like an interesting service, though!

JacobJans 12y ago

This is absolutely horrible and shows complete lack of respect for copyright. It should be taken down immediately.

Houshalter 12y ago
This isn't qualitatively different than readability or numerous other extensions. The only difference is he is doing it server side.
- declan 12y ago
  > The only difference is he is doing it server side Ah, but from a legal perspective, this makes all the difference. It is, I'm sorry to say, copyright infringement and goes beyond what fair use permits.
  - petercooper 12y ago
    You're right, but I wonder where the line is drawn with things like online RSS readers. They're basically doing the same thing, but just behind a username/password.
    - declan 12y ago
      >line is drawn with things like online RSS readers For starters, I'd look at the Supreme Court's opinion in Grokster. Here it is in HN-suitable pseudocode. :) IF $distributing_product_used_by_others_to_infringe == FALSE THEN FIND_NO_LIABILITY ( "None found" ) ELSEIF $distributing_with_object_of_promoting_infringement == TRUE THEN FIND_LIABILITY ( "You're inducing infringement ") ELSE IF $actual_knowledge_of_infringements == TRUE and $failed_to_block == TRUE THEN FIND_LIABILITY ( "You're a contributory infringer ") ELSE IF $capable_of_substantial_noninfringing_uses == TRUE THEN FIND_NO_LIABILITY ( "None found" ) ELSE FIND_LIABILITY ( "You're a contributory infringer ") END END END How this applies to online RSS readers is left as an exercise for the reader... UPDATE: A copyright lawyer friend points out via email that I neglected to include vicarious liability. I'll leave the above pseudocode as-is and note vicarious liability here as a bugfix/todo.
  - Houshalter 12y ago
    Yes I understand that, I said it isn't qualitatively different. If the web page somehow distributed javascript that would do the same thing client side, the website would appear exactly the same, yet would be perfectly legal. The comment I was replying to wasn't arguing that it was illegal, but actually wrong and "lacking respect". I was just pointing out that he should have the same opinion towards things like readability. Also I think readability did do it server side, and so do other things like RSS readers.
    - declan 12y ago
      Without taking a position on Readability, there's more to the calculation than client side vs. server side. For more see my post here: https://news.ycombinator.com/item?id=8455648 https://news.ycombinator.com/item?id=8455648

jawns 12y ago

Three criticisms, in descending order of importance: 1) The length of these excerpts goes a bit beyond what is typically considered fair use. You're pretty much reprinting the entire articles. So you may run into copyright infringement claims. 2) It looks like you're scraping the content, and your scraper isn't up to the task. One of the excerpts reads: "Your email has been sent. An error has occured and your email has not been sent. Please try again. • You can't enter more than 20 emails. • You must enter the verification code below to send. • Invalid entry: Please type the verification code again." 3) Borrowing the multicolumn text layout predominant in print is great when the columns all fit on the screen, but I found myself having to constantly scroll up-down-up-down-up-down to read these. So, in sum, you're only presenting the top stories successfully some of the time; when you are successful, you're infringing on copyrights; and the layout is not actually cleaned for readability.

btbuildem 12y ago

Speaking of readability, have you heard of https://www.readability.com/ https://www.readability.com/ ? Works wonders on all but the most stubborn webpages.

eli 12y ago
http://www.diffbot.com/ http://www.diffbot.com/ offers this service as an API

JustThrowinAway 12y ago

My first thought was how these articles were in no way more readable than the originals. Does anyone prefer a multi-column layout?

singlow 12y ago
Multicolumn is fine as long as the vertical height is less than the browser window. For long text this is really hard to make work in the browser unless you take over the pagination with javascript. This is not a good example of using multicolumn text.

andremendes 12y ago

http://getpocket.com http://getpocket.com do this "cleanse for readability" to the saved links but I think they got closer to readable than you did. If you get to avoid copyright infringement, and get rid of this multi-column layout, I'm sure it would be better.

dotsam 12y ago

The generated podcast is also presumably a violation of copyright (i.e. it creates a derivative work).

Xeoncross 12y ago

Neat idea, I'm sure you are violating fair use - but it is a neat little mashup of scrapers and text-to-speech. I won't ever use it though. I have a bookmark for cleaning up pages.

shmapf 12y ago

For those wondering about the usefulness of this. It should be great for those who commute on the tube or somewhere else where there is no phone signal. This lets you load a single lightweight page before boarding, and then be able to browse hn easily. Would be ideal if it was sent out in an email though so you don't have to remember to load the page every time.

lziz 12y ago

Cleaned for readability, eh? - http://i.imgur.com/faqQrQi.png http://i.imgur.com/faqQrQi.png

kolev 12y ago

I think a script that does this, but puts in your Pocket with an HN label would be even better - you keep track of what's read and what's not as a bonus.