Linenum - Curated tech and maker articles

I got tired of reading what’s trending on HN and Reddit. I thought that there must be way more interesting content that isn’t making it to the front page that I was missing out on. Boy was I right! I built a couple of automated scraping kits and found independent developer blogs through the Github API and indexed all the content in Algolia.

Some search terms that might bring back interesting content: burnout, depression, launching your product, etc.

This is a soft launch and I plan on doing a PH launch soon. Any feedback would be awesome.

Check it out: https://www.linenum.com/

3 Likes

Thanks for sharing your project @nav! It’s true there is a lot of great independent content that never makes it to the top of HN/Reddit. I could see this being a good way to find articles that might not be trending but are highly relevant to your company, tech stack, interests, etc.

Here’s some feedback I had after trying it out.

  • I’d recommend highlighting search terms in the results. This is important to helping the user understand why a particular result is being shown. Our SDKs and InstantSearch library support highlighting out of the box, let me know what you’re using and I can point you in the right direction.

  • I liked the Explore search UI (the one in the screenshot below) better than the search on the home page. With the horizontal results it’s hard for me to know what was the most relevant result, the 2nd most, 3rd most etc. Having two different interfaces also risks potential confusion about if the search itself is actually different between the two, or in this case if it is the same search presented two different ways.

  • I spent time trying to figure out what the green check mark did. It seems like it’s an indicator of when the search is complete? If that’s the case I don’t think it’s necessary because the hits themselves will indicate that the search has returned.

  • The Algolia logo should link to algolia.com. We’ve refreshed the logo too, here’s a new asset you can use:

Let me know if you have any questions. We’ll be happy to help you promote on PH once it’s ready!

3 Likes

@dzello Dude! Thanks a bunch for the really thorough feedback. Much appreciated! :tada:

I made a couple of updates based on your recommendations. I wanted to start off by saying that highlight article is pretty awesome and it’s great to see how the highlighter works under the hood. I went ahead and implemented it - can’t believe it slipped my mind.

For the home vs explore pages, I was attempting to create an interactive landing page. But that obviously didn’t materialize as I had expected :sweat_smile:. I did update it to just return some results stored in algolia and removed the search portion from it. So now you can only really search on the explore page.

Thanks again, working w/ algolia is pretty straight forward and the docs are sane, so I was able to knock this out pretty quickly.

In regards to PH, that would be awesome. I’ll take all the help I can get!

2 Likes

Happy to help! The changes look great.

Hey @nav, good job on this project!

I just tried and really liked the results and some articles I found with it :slight_smile:

A couple of feedback regarding minor possible improvements:

  • The “Load more” button is a bit misleading in case there are no results
  • Have you considered using “remove words if no results” option? It may reduce relevance in some cases but it may also bring some interesting results when a very specific query doesn’t match anything

These 2 points would address the following result I got:

If you’re looking for the option I mentioned, it’s in the Ranking > Query Expansion section (but you can also set it at query time).

These are really minor improvements but I really liked the project so I thought it was worth giving the feedback :wink:

Happy to help more as needed! Keep up the good work :thumbsup:

Cheers,
Julien

2 Likes

Thanks @julien.paroche @dzello

I enabled query expansion to see how the relevancy drops when no results are returned and it definitely seems helpful from a UX perspective. I’m going to keep that on and play with the other settings and see what’s best for this use case. So thanks for the tip!

Otherwise regarding the feature set, if you were to use a tool like this, what kind of features would you like to see in addition to search? I was thinking of adding a ‘reader view’ - kind of like instapaper or safari’s version to give a user the ability to view the content consistently. Although it originates from many different sites. Kind of brings up an attribution issue though…

1 Like

Allow me to drop a very quick suggestion here, which is to disable native autocomplete on the search input to avoid getting this:

Simply add autocomplete="off" to the search input should do the trick.

Looks great :wink:

1 Like

I would be interested to see if there were any tweets or conversation about each article without having to jump off to twitter.

1 Like

Really nice udpates :slight_smile:

I like the idea of @dzello to have external metrics from other services associated with results.

About the “reader view”, it sounds great to me as it would allow the user to quickly check a bunch of different pieces of content without having to actually open the web pages. I’ve heard about the Mercury Parser API as an alternative to Readability. Maybe that could work if the “attribution” is not a big issue?

Thanks all!

@julien.paroche @rayrutjes @dzello Thanks for the further feedback. And ray I’m definitely going to remove autocomplete. I can see how that’ll become an issue for some users.

In regards to pulling in extra conversation related to a post (comments, tweets, etc) I can definitely see how that’ll be helpful (and also monumental :neutral_face:) but I like it. It’s something I’m going to have to think about building reliably…

In regards to the mercury web parser API, I have taken a look at it in the past and it definitely looks awesome but I believe linenum goes against its TOS as stated by the following clause:


"cannot extract content, either individually or en masse, for the purpose of republishing, emailing or transmitting such content on the Web including but not limited to the purpose of misappropriating such content or in any way displaying such content as your own, unless you are the copyright holder of such content. Extract content for the purpose of publicly republishing it on the web, unless you are the copyright holder of such content."

I already have all indexed posts’ content though, through some fancy heuristics on my side. So if it’s something everyone is interested in, I definitely have all the data to do it!

1 Like