Practical Developers,
I write to you in an hour of great need. I have, in the parlance of today's internet, done a thing, and might for the very first time in my life "ship" said thing. Unfortunately I have a problem I cannot tackle on my own. I need testers.
In order to give background, I have to confess something: I am a tab hoarder. Or at least I was, until I built LinkMeLater. A few months ago, my friend Jared brought up a weird idiosyncratic behavior he noticed a few people exhibiting: emailing links to themselves for later consumption. I mentioned that I actually did that too, and that I would often go through my 40+ open tabs in a window, close the tabs I didn't need anymore, and email myself the articles or videos I wanted to get back to at a later date. He argued that that whole process really shouldn't exist, given that there are both plenty of tools for reading-list management, tab management, and link sharing.
Over the course of the conversation I noticed that the key similarity between all these services was that they expected you to adopt new, sometimes unfamiliar behavior, for a marginal payoff at best. It's much harder to remember to hit the Pocket button and then remember to read your Pocket collection (even if they remind you), than it is to just email it to yourself and only zero out your inbox once you have time to check out whatever you were looking at.
The LinkMeLater service then picks up any link saved more than 1 day ago, and randomly decides whether to send it to you. These semi-daily emails come either in the form of a series of individual emails for the day, or a single digest email that contains all of them.
Here's my problem though: I would love to launch this to a larger audience, but I have only managed to get 3 friends to test this. I'm not sure what I should add to it next, or where improvements need to be made. The tool fits my needs, but I would love to keep working on it. My current roadmap includes weekly/monthly/user-scheduled emails (as opposed to semi-daily), user-controlled link selection probability, and additional delivery media, such as SMS, but prioritizing those things means figuring out what a potential user wants most.
If LinkMeLater sounds useful to you, I'd love it if members of this great community checked it out and gave me feedback. To set it up, simply sign up on the site, and download the chrome extension that's linked on the homepage. Thanks for reading, I hope to hear from some of y'all π.
Edit: Thanks to some comments from readers, I gave porting LML to Firefox a shot, and it's now available at https://addons.mozilla.org/en-US/firefox/addon/linkmelater/ π.
Top comments (21)
Ironically, I need this saved to link me later :)
Ha, I'll put "recursive promotion" on the roadmap!
I can relate to this.
I currently have 100+ browser tabs on my phone and who knows know many saved articles in Pocket.
I'll be downloading this and testing.
Glad to hear it resonates, Shannon! Thanks for helping out :).
I'm happy to help! I just downloaded the extension and have an initial comment.
On the dashboard, it would be helpful if it showed which email address I have associated with my account. I signed up without a problem, got the TabMailer Activation email (which came SUPER fast, btw. Kudos!) but without having that email handy, I wouldn't know which one was associated with my account. I would also assume, adding another email address in the "Enter new email" field overwrites the previous one (as opposed to emailing both). Is that correct?
It's not a major thing, but I could it see it being a pain point if a user put in their email address incorrectly on set up and wasn't aware that it was mistyped.
That's a great idea, I hadn't thought of that, thanks! Should be a fairly simple addition.
Yes, it does overwrite the previously associated email, I'll be sure to add some copy for that when I add the above suggestion. The writing will most likely go through quite a bit of change over the next few weeks, much of it was written at 3 AM as I kept plugging along on the project π .
Good to hear about the email, I was having some issues early on with Mailgun emails being marked as spam by default.
Thanks for the thoughtful feedback!
I knew that I'd have to/want to add other OAuth providers down the line, but in testing with friends it's been a much more frequent request than I thought. I am afraid that I've coupled tightly with Google Auth (thinking it was a Chrome Extension after all), but it should be doable with only minor re-work to the user flows (around the automatic login into the extension). I'll comment back here once I am underway. Luckily the approach I took involved backflips to avoid logging the user into Chrome, the side-effect of which may be that this may be slightly easier to do in the extension portion.
On the subject of WebExtensions, I was aware of them when I started, but the LML does rely a bit heavily on the Chrome APIs. I will take a look at the portability subject :).
Iβd be interested in Firefox support too.
Mozilla provides a nifty automated compatibility test as a first step to getting everything compatible. It threw a weird report when I submitted the extension (The "Report" section where errors should be is just blank), so I am reaching out to see if I can get a response before I dive deeper.
LinkMeLater is now available for Firefox on AMO at addons.mozilla.org/en-US/firefox/a...
Unfortunately doesn't seem to work - it loads a tab to localhost:5000, which doesn't seem to work.
Hrm, looks like maybe the way manifests are generated differs between the two? That should only occur when it is running unpacked (ie when I am working on it locally). Thanks for bringing it to my attention, I should've thought to test that, I will upload a revision as soon as I have a chance to look into it.
I am working on getting this fixed, but I am blocked by AMO/Firefox Add-on Review. It looks like for listed extensions they disallow any remote scripts (I use apis.google.com for auth), so I may have to figure out an alternative distribution method once I get the fix for the localhost thing (which is just a difference in how the browser can be detected between chrome and FF).
Happy to test but I avoid authorizing via Google. Email + pwd sign in please, if you get a chance.
Whoops, totally missed this.
I am doing my best to work on adding other Auth providers in (first Gitlab, then others if I can refactor sufficiently), but I don't see myself implementing email + pw any time soon. It would require more work than simply adding additional providers, would alter flows, and probably require gathering (and managing) more data about the user. I understand the concern though, I have been meaning to get around to writing a privacy policy/breakdown regarding what the app actually gets from Auth.
The project as a whole is on-going, I have actually just finally added support for other auth providers like people were asking (twitter is still WIP because they took two weeks to approve my request), but the FF version got taken down because they generally don't allow some option I needed, I think it was requesting a remote script for google auth. now that that's fixed I will look into re-uploading to Mozilla.
If anyone has feedback, feel free to post it here or send it to me at admin@linkmelater.win :) .
I use a tab manager and have more than 1000 in there, saved indefinitely until I may or may not use them again
I used something called Toby for a while that is a little bit more in line with a traditional bookmarking/tab management thing, if you haven't tried it it's pretty cool.
Not sure who gets notifications on what but check out LinkMeLater for Firefox at addons.mozilla.org/en-US/firefox/a... π
Hi - looks like this is no longer available?