4 ms·
Opera's UNIX dev on how he uses wget and grep to read blog comments
- sucuri2 16y agoFunny to see an Opera guy suggesting a way to browse the web without using Opera :) He even says it is easier to read all the comments that way (via text using curl / wget/ grep) instead of using their browser (not that I don't agree with him).
- earnubs 16y agoI'd say it's more a criticism of the design of comment threads than the browser.
- DisposaBoy 16y agoIf that's the case, shouldn't they/he be adding some functionality via plugin(or built-in) that improves the situation in Opera.
- burgerbrain 16y agoWhy would you do that, if using wget and grep were easier?
- nl 16y agoBecause if he needs to do that, then presumably other people have the same problem, and I'd be reasonably sure at least some of them can't or aren't prepared to use wget, sed & grep to find things they did in another program. Browsers already have history, so it seems to me it would be logical to extend that functionality to show how a page changed in response to an input somewhere in your history.
- DisposaBoy 16y agoIt's not easier. It's just easier than what we already have.
- nettdata 16y agoYeah, but he's probably one of those guys that's still using Pine for his email.
- famrey 16y agoHe isn't suggesting a way to browse the web without Opera. He's suggesting a way to overcome design problems on a site.
- vdm 16y agoTake away: including permalinks for included bits of pages (such as comments) can be useful when they are crawled. Encouraging serendipity.
- jlees 16y agoOr just allowing a 'view all comments' option that could be easily Ctrl-Fd.
- katovatzschyn 16y agoRuario maintains the Opera AUR entry for Arch Linux also, and does very good job of it so earns credit: http://aur.archlinux.org/packages.php?ID=10333 http://aur.archlinux.org/packages.php?ID=10333 http://aur.archlinux.org/packages.php?ID=33276 http://aur.archlinux.org/packages.php?ID=33276
- sfphotoarts 16y agoTalk about wrong tools for the job. A far simpler to write, and more robust solution could be had many other way, simple_html in php for example, or javascript, any DOM parser will be more robust and easier to follow than all those sed/grep calls.
- oconnore 16y agoI don't think there is anything simple about learning PHP to find a few blog comments, or learning to use a 3rd party DOM parser when you are already familiar with tools that apparently work for this one-off application.
- xtacy 16y agoHe's not alone. :-) Richard Stallman goes a step further as his laptop isn't directly connected to the internet. Quote: For personal reasons, I do not browse the web from my computer. (I also have not net connection much of the time.) To look at page I send mail to a demon which runs wget and mails the page back to me. It is very efficient use of my time, but it is slow in real time. Source: http://lwn.net/Articles/262570/ http://lwn.net/Articles/262570/
- zacharyvoase 16y agoGreat approach, but some work could have been saved (and robustness added) using the W3C’s HTML-XML-utils. For example, there’s hxselect, which filters HTML/XML against a CSS selector, and hxpipe, which breaks XML input into a more grep/awk-friendly format. I’ve used these tools myself on multiple occasions, they’ve saved me a huge amount of time.