Will I implement at least 15 changes to my search feature by the end of October?
18
256
350
resolved Oct 27
Resolved
YES

People seem to like my new search page. I'd be interested in making it better, and would appreciate suggestions.

This market resolves based on what I actually impliment, not just on how many things are suggested. I am more likely to impliment changes that are:

  • Useful for practical purposes.

  • Wanted by more than one person.

  • Easy to impliment.

That's not to say I won't take on harder or more niche projects, espetially if it's something I myself would find useful. I enjoy programming, I just don't have infinite time. Please suggest even minor improvements that you'd like to see; the worst case scenario is just that I decide not to do that.

If you see a suggestion in the comments that you'd also find useful yourself, please like that comment or indicate your agreement in some other way, so that I can get an idea of what to prioritize. (I've also commented some ideas of my own; please like the ones that you'd find useful.)

I won't bet.

List of new features that qualify for the count so far:

  • Added a way to sort by creation date.

  • Added a way to search for markets created by a specific user.

  • Secret feature.

  • Added the ability to search multiple-choice answers.

  • Reduced bandwidth usage by recreating market links from the slug, and added my referral link to the end as a costless way to support development.

  • Added a dashboard feature.

  • Added the ability to search by a market's groups.

  • Added a "clear all search fields" button.

  • Added a way to bet directly from the search page.

  • Market data now displays in a grid, and you can choose what information to see.

  • Markets you interact with (click, bet, or add to dashboard) now update much faster.

  • Added a random sort option.

  • Added filters for all numerical market properties: close time, creation time, probability, liquidity, and number of groups.

  • Added a filter and sort option for the last time a market was updated.

  • You can click a creator's name to see only their markets.

Get Ṁ600 play money

🏅 Top traders

#NameTotal profit
1Ṁ114
2Ṁ92
3Ṁ88
4Ṁ32
5Ṁ32
Sort by:

Thanks for all the suggestions! Next month's market here:

predicted YES

I think writing dates to the Close time field does not work. I tried the stuff from the documentation (at least "1m", "1w", "11/15/23", "11/15/2023", "15/11/2023", "11/15", "15/11"), but all I get are mysterious mangos and sassy strawberries.

@Loppukilpailija Are you only entering a single time? If so you're only going to get markets that match that exact time, which is rather unlikely. You probably want to enter a range, such that it can match any market within that range.

predicted YES

@IsaacKing Ah, of course, didn't realize I need a range. Thank you!

bought Ṁ80 of YES

Resolves YES, right?

Updates from today:

  • You can click a creator's name to see only their markets.

  • Date entry for creationTime, closeTime, and lastUpdatedTime is dramatically improved. You can enter dates as yy/mm/dd (or just mm/dd for this year), or use relative timestamps like "1m" to mean "1 month ago".

  • Loading lag has been reduced. It will now only freeze the page for a few brief ~1 second intervals while markets are loading, rather than a solid block of ~10 seconds. Still a lot of improvements to be made here...

Filter by market type. For example, only display unlinked multiple choice markets.

Add support for other platforms. Polymarket, Kalshi, etc.

Update market data faster. Various improvements that could be made:

  • Retrieving new markets can happen concurrently rather than one at a time.

  • Use a market's lastUpdatedTime to determine how frequently it needs to be queried.

  • Listen to /bets and get a new probability for any market that was just bet in.

  • When a user places a bet via the search page, use the returned probability to update that market immediately.

  • Continue to pester Manifold until they fix all the various causes of lag and errors in their API.

  • Query markets more often if they're near their close date.

  • Get market data from supabase instead of the Manifold API.

  • The full dump of all markets doesn't need to be crawled that frequently, and can happen concurrently with other market updates rather than putting them all on hold.

Make the choice of which markets to store locally smarter, so it won't accidentally cross the 5MB localStorage limit.

Fix the live loading to not block user events for the first few seconds while markets are being loaded. Right now clicking on a market or editing a dropdown menu will do nothing, as the page is too busy in the background.

The loading indicator also updates sporadically rather than at set intervals due to this, which looks bad.

@IsaacKing This is partially fixed.

Add an option to display each answer on a multiple-choice market as its own row in the search results. This way people can easily find answers to bet on in large multiple-choice markets without having to scroll through and read the entire list on Manifold. And it will help get around the lag that one currently has to deal with to bet in a market like the speaker one.

Related: Let people select multiple answers and bet in them all at once, providing a single total amount of mana and a single target probability that's split up along them all in an efficient manner. (Not equal; answers further from the target probability should get more mana.)

Additional option: instead of a target probability, let people specify a target change, e.g. "increase probability by 1.5x". This way they can bet multiple answers down for arbitrage without taking a position on which is actually more likely.

Edit existing rows when there's a change to a market's data or the order of search results rather than re-rendering the entire list. This would make rendering faster, and fix the bug where the user's cursor will disappear from a certain row when the list is re-rendered.

I should also make dates update without needing to re-render the entire list, so that a seconds countdown to close remains accurate.

Include unlisted markets more consistently. Right now they'll only be included if they happened to be listed during some point my scraper was running and then got delisted later. Could be accomplished by checking the supabase API where unlisted markets are available. I could also let users directly submit unlisted markets that they're aware of into the search results.

bought Ṁ10 of YES

Suggestion: In addition to the "mixed" option, allow sharp cut-offs.

Example use case: I would want to get a list of all markets that resolve within the following 30 days, sorted based on their probabilities. I don't think I can do this currently. (I can give weight 1 to both close time and probability, but that gives a quite unstructured list.)

I've added a faster check for markets that you interact with. If you add a market to your dashboard or place a bet on it, changes to it will be reflected within seconds rather than minutes. This should make live trading much more viable.

@jacksonpolack Suggests hotkeys for quickly placing bets or limit orders. Not sure how you'd select the appropriate market from the list, hmm.

@Joshua and @Ernie were interested in a dashboard for live events. Users provide a list of markets, and the site displays the markets' probabilities in real time and lets the user place bets on them from the dashboard.

@IsaacKing I've added a very lightweight dashboard feature, I'm going to add more to it over the next few days. Please suggest anything you'd like to see. We're also talking about it on Discord here.

You can now place bets directly from the search page.

Multiple-choice answers can now be searched along with the question and description.

@jacksonpolack suggests a way to search by recently subsidized.