Integrating websites with the browser

There should be APIs for integrating websites with the browser. For example, having a browser-powered “save” mechanism, that the browser can display however it wants to, and can in theory integrate with the OS on mobile - imagine searching for something on your phone and being shown a relevant reddit post you saved a year ago; this is currently not possible, but it could be possible if websites could integrate with the browser. A different but similar option would be to have a “save” HTML tag, making saving explicit and inherently user-initiated, but websites cannot sync these across browsers and instead you have to use the browser’s built-in sync features.

Anyway, I digress. So currently, we have “desktop notifications”. It’s a bit of a misnomer as they also work on mobile, but the idea is that they help integrate websites with the browser. Websites can also declare search engines to be used by the browser. Why not go a few steps further?

This would significantly improve user experience for most web users.

Could you elaborate on the concept of “searching for something on your phone”? I’m not sure what that means. Do mobile OSes have something like a global search field for performing text searches across apps?

They do! They can search contacts, pictures, downloads, emails, etc.

I guess Windows 10 and macOS (desktop OSes) also have that feature, but as someone who uses Linux I’m not very up-to-date on that stuff.

But aren’t those in-app searches? The user can search for a name or email inside different apps, but they only get results from that app, not other apps.

On desktop, it’s a different story. On Windows, the Start menu has a search field, and on Mac, there is Spotlight, which can likewise find all kinds of things.

Apparently some Linux things also integrate with the Desktop Environment’s search system, depending on the DE. (There doesn’t seem to be a standardized way to do it yet.)

From what I remember, Apple’s iOS search field is basically Spotlight. Not sure about other OSes.

Anyway, this is going a bit off-topic. The thread is about integrating websites with the browser, not the browser with OSes.

Ah yes. I remembered where the search field is on iOS. I have never used it in the 3 years I own an iPhone. Apparently, it is integrated with Safari, so, for instance, browsing history is displayed in the results.

How does “saving” web pages relate to bookmarking them? What’s the added functionality?

I’m not sure about the bookmarks part, but my impression is that the original poster is asking for web standardization of something like iOS universal links, which integrate with the iOS Search index, as well as Android app links, which in turn integrate with the Google Search app index. If these ever will be standardized, they probably would be dependent on existing efforts in the Web App Manifest and Web Share standards; if the original poster is not yet aware of them, they should check them out.

No, I’m not looking for app links or universal links or whatever it is because that seems like just yet another ad mechanism of sorts. (visit site, get ads for site? yuck)

I’m talking about searching the user’s own content: the user’s profile, the user’s saved/bookmarked posts, the user’s own posts, boosted posts, and so on. Things that require you to be logged in, not things anyone can access.

I think most of this feature can be done from a web browser address bar - for example,
when I search for my username, it shows me things from my content like profile page,
posts and some private pages that were added to browser history after logging in.

You can probably find that content with the search area on a mobile device if it includes
web browser history. Also, the app/universal links idea on Android / iOS, that can
"Seamlessly link to content inside your app", could only be useful for web apps.

• Bookmarks: you can save the URL of a webpage for quick, simple access.
• Saving a webpage (e.g. “reading list”): you can download it to use offline.

There’s an idea from the original post user about this:

Maybe there could be an API for this that gets the browser to ask the user
if they would like to download the webpage to their device for offline use,
something like “discourse.wicg.io would like to save this page to your reading list”.

“Maybe there could be an API for this that gets the browser to ask the user if they would like to download the webpage to their device for offline use, something like “discourse.wicg.io would like to save this page to your reading list.”

It is already possible for sites or pages to save themselves for offline with service workers and the cache api.

1. Are those searchable? (can the user search them?)
2. Are those filterable? (can the user choose which ones get searched?)

With, e.g. bookmarks, the user chooses which things get bookmarked.

this can be done with the OS not the browser

The browser should be able to automatically import bookmarks from reddit/twitter/etc whenever the user wants to. The browser should have an API that allows those web services to easily export their bookmarks.

This is about interoperability and portability.

I may misunderstood your main goal, if you mean when a user of a device searches with some key wrods, the browser provides some suggestions? but the search input box is within the OS of that device, the OS can choose to show some suggestions, not the browser, even if a browser tried to, but the OS didn’t let that certain browser to communicate with the main search box in the OS, it will be useless

No, that’s not it either.

Consider these:

• About Twitter Bookmarks
• How do I save a post and where does it get saved? – Reddit Help

What if these were easily synced with the browser? What if the browser had a button that showed these consistently across websites, as title-url pairs, similar to browser-level bookmarks of today?

What if the browser pulled the user’s data so if the user gets banned, the browser has it all there for them?

the last one gave me the clear idea of what you want, and is great to be included in the browser, I and my team may build a new browser (and I may include this too) if you want to contribute with us, you will be very welcomed… ! the schedule is yet not decided for, but will let you know

Hi all! @simevidas @SoniEx2 @hazho @DanielHerr @jschoi @Edward_J

1. context

1. What you’re talking about looks like RSS - Really Simple Syndication: RSS - Wikipedia - not all sites support
2. if that’s what I’m thinking… in this case it’s RSS - I would find it interesting if the browser has a previous option to integrate websites inside the browser through an api that previously captures what I want - reference here: GitHub - taroved/pol: RSS generator website - Pol is RSS generator website
3. It should be possible to configure or generate an rss from any site and with that I can receive notifications of the pages that I would like to have notifications but does not support RSS

2. concept

image1365×328 13.4 KB

2.1 image description

1. As we can see in the image, I can receive notifications of the website contents that I need through the browser itself, even when these pages do not have access to RSS - Really Simple Syndication or web mention
2. This happens because the RSS - Really Simple Syndication is done locally, I just add the page address and if the page doesn’t change the address I will continue to receive my new content notifications -
3. Not all browsers support RSS
4. A partial way to solve this problem would be to use a cdn-p2p network, so the content is distributed among different people and you can receive updates from several sites at the same time without losing speed in terms of page loading
5. If these two solutions: cdn-p2p and rss are bad ways to integrate sites in browsers - the last way that is over and is less viable I think… would be to use IPFS which is web 3.0 network protocol, it works as if it were an rss/cdn-p2p - that is, you distribute the content of the pages and receive notification when this content is modified - the last way… IPFS is a mix of rss/cdn-p2p from what I know partially

3. References

• Generate RSS feeds for any web page | PolitePol
• RSS - Wikipedia
• https://feed43.com/
• https://rss.app/
• Website Traffic - Check and Analyze Any Website | Similarweb
• https://fetchrss.com/

Maybe web bundles can help with this specification here