Hacker News Reader

Comment by sentenza | Hacker News Reader

The promise was that, as search engines get better, there would be no need for bookmarking and downloading, since all the content in the world is only one sloppy search query away.

Somehow, I seem to have ignored this stubbornly and instead bookmarked pages, downloaded videos and copied strings (!) out of the web for years.

Not that it did me much good, I now have thousands of inconsistently sorted bookmarks and a bunch of directories filled with assorted data.

What this article tells me (again), is that because you can't trust search engines to behave in a predictable manner, you can't even be sure you'll be able to find the stuff that you know is online.

View on HN

I've found that copying recipes off of webpages is the most important one - you'll never find the same recipe twice.

That is quite interesting.

Any idea why?

Several reasons. There is a seemingly unlimited variation of each basic recipe name (often the same exact recipe named multiple ways; or slightly different recipes named the same). There are thousands of recipe sites and blogs, and ten million unique recipes. The popularity of recipes and recipe sites can change rapidly, and often does; due to seasonality, food trends, seo, competition - so that recipe you found at result #4 on Google, isn't likely to be there a year from now.

Online recipes is herding cats.

Are you sure it's because of the website change instead of the search engine dynamics? The search engine results are not organized, therefore it's unpredictable where a particular results may show up. If the website changes it's keyword combination in tags on the page, it may also disappear from our search results using the same keywords as before.

So we should save it in categories instead of using keywords only. Try the following link and search for "recipe" than click "more similar links..." may give you some idea of how to save an organize your useful information into your private web. (This site is in early beta, there are not much data available yet.)

http://usweb.bingobo.com/usweb/welcome.jsp

I typically use google images to search for recipes, I very rarely have issues relocating a recipe.

Similarly, I found a page a few days ago about fixing the rMBP trackpad in Windows 8, and now that I have Windows 8, I just can't find it again! Really wish I'd bookmarked it.

In Firefox, you can roll back through your history for several days of links, and search the link text. Does your browser keep a history?

You can't really trust bookmarks either. Most of my bookmarks are dead either because the site is now gone or because some blog author decided to change their path format and didn't bother setup 301 redirects.

Microsoft's help pages are the worst for this. "Oooh, confusing .NET problem I'll google a response - Hey, a nice forum-post explaining that MS has a document telling me how to handle this issue click NOOOOOOOOOOOOO".

Whenever I have a C# problem and google it, I always skip the MSFT documentation and go straight to any possible Stackoverflow or discussion page.

Same here, it's not even worth going on their site.

There used to be a whole slew of helpful msdn (technet?) blog posts which now not only redirect but also require logging in with an MS account to read them.

It's really unhelpful.

edit: Actually I just checked and it looks like they've eased up on the need for logging in, I guess they saw too much of a backlash. (Or just flip flopped their own decision again.)

I started saving/clipping entire articles for this purpose. Evernote has a 'clipper' functionality which works fairly well, or you can print to PDF.

I tried Evernote. I can archive information pages into notebooks as groups and find by tags. But it cannot save dynamic product/service information with updates, and it's not self-organized. So like other online bookmarks, it's not popularly used.

That's what the Internet Archive is for :)

And there's an interesting idea for a browser extension.

If the page the user is looking for no longer exists, check the internet archive and present that instead.

I use this https://addons.mozilla.org/en-US/firefox/addon/resurrect-pag... It's not automatic, but I guess it wouldn't be hard.

I pay pinboard.in to host cached copies of everything I bookmark.

This is exactly right and the sole reason I use Evernote for its full page clipping abilities.

For keeping local copies, https://addons.mozilla.org/en-US/firefox/addon/mozilla-archi... is a good add-on that improves Firefox's save functionality. I use it very frequently: ctrl-s, Enter, and I get an .mht of the page I'm reading.

For remote copies, https://archive.org/ can be instructed to keep a copy of a page, e.g.

https://web.archive.org/web/*/http://news.ycombinator.com/ to check for copies

https://web.archive.org/save/https://news.ycombinator.com/ to record a copy

These bookmarklets can be used to go from your current page to /web/ or /save/:

  javascript:(function(){window.open('https://web.archive.org/web/*/'+(''+window.location));})();

  javascript:(function(){window.open('https://web.archive.org/save/'+(''+window.location));})();

Unfortunately, this won't work on sites that robots.txt-block Internet Archive, and existing archives may be made inaccessible by a future robots.txt block.

http://archive.is/ is another useful site that executes all of the JavaScript and captures the post-JS-execution DOM.

Thank you for your information. I'm looking for tools to save my useful information like other people do. I have two problems with this type of archival tool besides the robots.txt issue:

1. The static snapshot can only give us an idea of what it looked like because it's not functioning, not updating. Better than nothing, but we may have no clue what we want to find other than a few keywords. If you need to search the entire context of the archival, everybody needs to install a mini search engine.

2. If it becomes popular like Google, so many people are going to use it, it will create so much data duplication on everybody's pc, which is helpful but not an efficient solution.

Do you like to have a self-organized personal private web on top of the public web via a few clicks? http://bit.ly/1c2DzN6

You have described exactly what I use instapaper for.

It is a crying shame that the instapaper search is nearly useless, but it means I can usually cut down my search set to my own 500 saved pages vs the entire web.

>The promise was that, as search engines get better, there would be no need for bookmarking and downloading, since all the content in the world is only one sloppy search query away.

That's exactly why you need something like discussions: refinement. There's no specific 'right result' for every possible user and query. Google might give precedence to blog posts or articles from big sites for some queries but what I really want is to search forum posts. They might as well just remove the search options for images with that premise, since they supposedly know what you really want and you don't need to filter.

"The promise was that, as search engines get better, there would be no need for bookmarking and downloading, since all the content in the world is only one sloppy search query away."

It was back to the AltaVista age, when the web is simple and clean. Just like you, I could never think we can keep using keyword combinations to filter the search results to fit our needs forever.

We need a self-organized storage of information, we need more efficient ways than keyword combination, and we need the search results to be consistent in a period of time.

Can we achieve this goal and meet the users needs in the near future? Maybe: http://bit.ly/L6RKdN

I built my own little tag + bookmark browser extension with the dropbox API (my app is not public). It works for me. Just gotta get better at tagging. Basic browser bookmarks are not good enough.

Evernote

Reply on news.ycombinator.com