First public beta for Arc Editor!

Announcements
1

Hi everyone! So a little background first:

Arc Editor is a ground up rebuild of Arc Timeline app, and also the underlying recording engine called LocoKit. The reason for doing this is to get rid of the almost decade of cruft that’s built up in the Arc Timeline app and underlying LocoKit. Over the years the codebase has become more complex, and sometimes more challenging to work with, which has lead to more bugs, and more mysterious behaviours. As new versions of iOS arrive, the “best practice” techniques and tools have changed, and Arc has become layered with generations of those, feeding into that growing complexity and potential for things to break or go weird.

So in comes LocoKit2 and Arc Timeline Editor! Everything fresh and new, all shiny and clean and simple and easy to maintain and build on top of. Yay! :tada:

So far Arc Editor has the daily timeline view and timeline editing views all built out, largely complete. But other parts of the app are either missing entirely, or in the very early stages of development.

Timeline view doesn’t properly show weekly, monthly, or yearly views yet. Overview tab has been replaced with a Places tab, for gathering statistics about places you’ve been over the chosen time periods. The old Overview functionality is intended to be rolled into the timeline tab.

The Activity tab is partially built, but some of it is fake data, standing in as placeholders until the job of building it all out is done. It also has a lot of weird quirks, due to being quite incomplete code.

The Settings tab is currently a dumping ground for various debug tools and views, useful for the private and public beta periods. Over time I’ll flesh it out with more sensible stuff, like in Arc Timeline app. Though all those debug views, this time I’ll keep those in instead of hiding them. Lots of people find that stuff useful or interesting!

For feedback and testing: basically the goal is to get feedback on anything you find missing but important, or present but broken (and also important). Anything you want to do in the app that you either find doesn’t exist yet or isn’t complete enough to do. Anything you find that looks like it’s meant to be finished but is acting weird or poorly. That kind of thing!

I’ll be prioritising work based on the feedback from the public beta. So if there’s something you care about and it’s not there, or it’s there but kinda broken or incomplete, that’s important stuff to know, and I’ll focus on filling in those gaps first.

Now for how to get started with it: On the Settings tab at the bottom there’s a Database section. Open that up and choose “Old LocoKit database import”. That will start the import from your Arc Timeline app data. It’ll take 30-60 minutes or so (mine took 35 mins on my 15 Pro). You’ll probably want your phone plugged in during that time.

Once that’s done, all your Arc Timeline data will be in, and you’ll be able to browse around it in the timeline views and edit it as normal. You’ll want to keep both apps installed and running. Arc Editor isn’t finished enough yet to quite take over - there’s a bunch of things missing that you’ll probably still want, so you’ll want Arc Timeline around for a while yet.

The two apps record to separate databases, so they don’t share the recording role like Arc Timeline and Arc Recorder do. Eventually once things are far enough along, Arc Recorder will be changed to act as a backup for Arc Editor instead of old Arc Timeline. But for now Arc Timeline and Arc Recorder live in their world, and Arc Editor off in its own separate world / database.

Arc Editor doesn’t interfere with or interact with Arc Timeline at all. So there’s no risk involved in installing it. It’s siloed off in its own space.

Alright, time for the link!

The beta group is limited to a small number of people at the moment. But as long as nothing goes wrong in the first couple of days I’ll expand it out to a much higher limit over the following days. And I’ll likely be shipping updates quite frequently, based on the feedback here.

3 Likes
2

Installation and import of the old database completed in 10 minutes for 7 years of data (iphone 15 Pro).

The application is visibly more responsive than the old one. We’ll have to wait and see, but what I’ve already seen is encouraging.

Thanks Matt, for all your hard work.

1 Like
3

Great work Matt. Looks very nice. With only a few minutes of looking at it, just some things I noticed:

  • A minor detail: the import screen with red text (in dark mode) to emphasize do not close maybe is just temporary yet I think it could look better. Maybe a circular progress indicator would be bigger and easier to see?
  • Confirming visits: it seems that after confirmation, there is some process by which it does the confirmation and the UI can take some seconds to update, me thinking that something went wrong or the confirm didn’ go thru. Maybe it would be better to hide some row and then do the confirm with whatever db updates or other needed changes?
  • Font size: love that is now adjustable yet default being one (?) size larger than default I’m not sure about. I’m used to the font sizes in any list, table, or view such as Calendar, Music, etc. Even with that and then the default smaller size for secondary text such as travel, I think I prefer that rather than it being larger than normal. I am aware I can adjust it in a way yet if I want the smallest possible per iOS settings, that’s not possible. Seems that font size isn’t adjusting everywhere like such does in other apps such as Settings, Calendar, Music, etc. For me that is less important but for those concerned with accessibility such as vision impairment, you may wish to add.

I think I’m ok using both apps for now or may re-import if needed (or if there’s any change?) when final comes out. Any thoughts of if we should maintain both during this time or just use one or the other, with the beta perhaps just as a test for feedback?

That’s all so far. Looks nice and looking forward to updates and release.

fb2d7d-666b-e066-77c0-022182bbce51_Progress_View_In_SwiftUI
fb2d7d-666b-e066-77c0-022182bbce51_Progress_View_In_SwiftUI1280×720 32.9 KB

4

IMG_1281
IMG_12811179×772 67.9 KB

After first import and launch it seems there are some background tasks yet to be completed. I’m guessing maybe they are queued for some interval and thus one might need to wait some hours before completion? Not sure if as with import such should run and be verified on launch until such is ru at least once? Not sure but maybe such is related to the new total visits which for me shows 0 so far for any selected visits?

Naming: maybe this will be named Arc Timeline 4.0 upon release? Seems better.

Related to such, I like the app very much and continue to use it. Yet the thought came of what other popular apps are there and what might help Arc become more popular? Asking AI, found some examples such as Life Cycle, and PolarSteps Travel.

They may not overlap yet if some apps have thousands of ratings, maybe they have interesting features that might give you some interesting ideas.

5

Oh wow, that’s much faster than mine was on a 15 Pro! Mine is a bit over 9 years of data, and took about 35 mins each time I tested it. But I suspect my database has a lot of cruft in it, from the many years of using it as my main testing ground during development.

Yeah that’s my favourite thing so far. Well, and the extra data accuracy and detail over the old app. You’ll start to notice that in a few days, I imagine.

It’s much faster, processing almost instantly a lot of the time. And much more stable - there’s no known crashes, and the code is very clean and simple, as compared to the old jungle that’d built up in Arc Timeline app.

The improved recording detail and accuracy comes from the rebuilt LocoKit2, with better filtering algorithms etc. It also now uses a system of recording at full detail always, then pruning out detail later, as a deferred process.

Arc Timeline / old LocoKit would adjust its recording / sampling rate in real time, to reduce energy use. Which meant less detail up front. But phones are much faster and more powerful (and have more battery life) these days, so it’s now viable to record at full detail. Then later Arc Editor will prune out unnecessary samples inside Visits, and do “path simplification” on Trip items, using sensible simplification algorithms to retain the detail while removing unnecessary extra samples (eg a straight line only needs two samples: the start and end, not every sample in the middle).

6

Ah, I only tested the importer during daytime, so I never saw it in Dark Mode. I’ll make a note to test that and improve the colours!

I’ll give it a try :+1:t3: I’ve also got to build a similar full screen cover for the JSON export and import buttons. So I’ll work out a style that works nicely for all of them.

Yep. That’s a common annoyance for me too. There is a slight moment (or longer moment, depending on processing workload) where the UI should be indicating “stuff is happening”. Right now there’s no indication that your action was successful at all, until potentially a second or two later, which is quite meh.

Oh almost all text is using the default font size (iOS’s body style, which is the default for any Text where you don’t specify a specific style). Where are you seeing larger text?

In most cases I’m using body as the row content then something like footer, caption, or callout for the subtext in second line of the row. There shouldn’t be any text other than headers using anything larger than body. Maybe a screenshot will help me to spot which one you mean.

It definitely should be! I’ve set a rule that no text in the app is allowed to use custom font sizes - it all has to be one of the built in iOS style settings. So that everything supports Dynamic Type automatically. If you’re seeing anywhere that isn’t adjusting as it should for Dynamic Type, definitely send screenshots!

For myself, I’m going to keep running both of them, simply because Arc Timeline still has a lot of functionality I still depend on. Like the properly working Activity tab, workout importing, weather importing, a few other things I’m forgetting.

Already I’m kind of ignoring Arc Timeline unless I need it, because its data quality is lower than Arc Editor’s. It’s more pleasant to look at and work with the higher quality data in Arc Editor. But yeah, until Arc Editor has all the features I want or depend on, I’ll personally be running both and paying attention to both.

For everyone else, I’d imagine it’ll be a similar decision. If Arc Editor does everything you want, then there won’t be much point in running both. Though I’d give it another month or so, for Arc Editor to settle in a bit more in the beta process. There’s always a slim chance it might be necessary to delete it and reinstall.

Though Arc Editor’s full database JSON export and import features work MUCH faster than old Arc Timeline’s! For my 9 year database I can do a full export in about 5 minutes and full reimport in maybe about 30-40 mins. That’s several times faster than the old app, and much more reliable (new database schema has a lot more sanity and integrity checking built in). So doing an export → delete app → reinstall → import isn’t anywhere near as painful as it used to be.

7

Yeah after the import there’ll be ActivityType models pending update, and Places too. They get marked as stale / needsUpdate, and scheduled background tasks work through them once iOS decides to let that task run.

Those tasks are running relatively efficiently at the moment, but it’s still up to iOS to decide when it wants to run them.

For the ActivityType models, they’re meant to be automatically rebuilt when needed, so that if the app needs the models for your current region and they’re not built yet, that’ll be done immediately. But that process seems to be not always firing as expected, so it’s on my todos to debug that further. Though they still all get one in the end during the scheduled background task.

Ditto for the Place updates - they’re meant to be run immediately if needed, but sometimes I’m seeing that not happen. And it’ll stand out the most immediately after import from old LocoKit, because there’ll be hundreds of Places with stats not yet built. That one should also be trivial for me to fix. I’ll make a note to prioritise looking into it soon, because it can be a bit of a yucky experience immediately after import.

Yep! It’s eventually going to be the new Arc Timeline app, with the old one being deprecated and eventually removed from the App Store. I’m just calling it Arc Editor for now because it lacks so many features of the old app. So for now it feels better to call it a sort of “companion” app rather than full replacement. Although… it’s not really a companion app, in the sense that it doesn’t write to the same database. Eh, anyway, yep, eventually it’ll be just called Arc Timeline!

If there’s any features in those apps that you particularly like, definitely let me know! They’re not directly overlapping in purpose, as you note, but if there’s anything in them that does sensibly overlap and that Arc is missing, it’s still worth looking into, for sure!

8

Concerning other apps, I am not sure since I haven’t used them. I just noticed other apps that may not do the same thing yet have thousands of ratings; and wondering what prevents Arc app from achieving such. Perhaps such is mentioned just in the hope that the app will be around for a long time. :slight_smile:

Of font size, you are right. My mistake. Perhaps I had different settings for system wide and had some apps customized to be different. As for where such isn’t used, seems to be in the Settings tab for any detail views such as Task Queues, Background Tasks, and Database. Such is ok every where else I’ve looked such as Places → Country → details (First Visit, Last Visit, etc.). Perhaps they should match yet if Settings is just debug info for the moment, not a big deal. Such seems to adjust a little if I made font size adjustments but doesn’t match the Places tab details.

Thank you very much Matt for the continued great effort on app updates and support.

1 Like
9

I wonder if it’s because Arc is more niche, and also more expensive. Because Arc targets the highest possible detail levels it tends to be less mass market. I think partly due to higher complexity but also due to higher energy use, and less tolerance to termination. Data gaps in Arc tend stand out much more than data gaps in the other apps, due to the higher level of detail around the gaps.

So I think a lot of people switch off with Arc, and instead go to a less demanding app, that gets them the basic level of detail they want. For example I think a lot of people are happy to put up with Google Timeline’s very low detail timelines. And that app/service doesn’t require always-on recording like Arc, so the requirements are much lower.

Also… cheaper. A lot of them are either free or very cheap. Whereas Arc is priced at the higher end (though I’ve managed to bring the price gradually down over the years, as the customer base has grown).

Oh Arc definitely isn’t going away! I think the only thing that will kill it is if Apple decide to build in similar functionality into their own apps or iOS. Basically as long as there’s a workable revenue stream, Arc will stay alive. After all, it’s the business that pays all my bills and keeps me alive. I can’t exactly give up on it :joy:

Ooh, you know I think Claude built a lot of that view, and might’ve broken the font rules. Or maybe… maybe I reused that view from Arc Recorder app, which might have had custom fixed font sizes. Hmm. I’ll look into it.

10

Oh another thought on why Life Cycle has a lot more ratings: it’s got a free tier at all. Like, Arc requires a free trial, then requires a subscription (or lifetime purchase). So that shrinks Arc’s total number of potential users down dramatically.

The reason I eventually went that way with Arc was because of the “freeloader effect”. Basically - and this won’t sound particularly nice - with free apps or apps that have a free tier there’s a large number of people who act entitled, demanding, and judgemental, but will never be willing to pay.

So basically you get the majority of feedback and customer support workload from people who will never pay a cent for the product. And who also often act the most entitled and demanding. It’s incredibly unpleasant, and also really not a good use of my limited time and sanity resources.

For those years while Arc still had a free tier (maybe the first 5 years) I had to put in a lot more customer support work, and it was a lot less pleasant, and often outright insulting and demeaning. People abusing me over email, making all sorts of demands and insults, and insisting “I will consider paying for it if you add X and Y”, while obviously never intending to pay at all.

For apps and services with dedicated customer support teams that’s more viable. They can afford to handle that extra workload and abuse (though I pity the customer support teams - it’s often a terrible job), and use it as word of mouth marketing in the hope of getting a larger paying audience.

Though it often doesn’t work out that way - it’s often better business to just cut off the free tier altogether. Industry stats tend to support the removal of free tiers, on the ground that while it results in less total users, it leads to more paying users.

Not a particularly pleasant or flattering description of the situation! But yeah… ultimately Arc might be more niche, with fewer users, because I made that harsh decision. But it also leads to me being able to pay my bills more easily, and suffering less abuse and demands from entitled freeloaders :grimacing:

11

Understood.

As for my main use in some sense as a log of where I’ve been, somethings that I think would help me:

Mentioned before, search showing some secondary text such as location (city) would be really helpful, like Calendar:

IMG_1290
IMG_12901179×1448 147 KB

In search or when viewing a day, long hold on any location has some menu options. At least for me, among most used that could be added to that menu: Show in Maps (that’s how it’s worded in Calendar), and perhaps show other visits.

Not sure about others but I use search a lot. Certain apps like Music and Photos have a search button at the bottom, at least with iOS 26 because they are so used. I think that would be nice and as Settings is perhaps less used, maybe that can be moved to the more (3 dot menu) for a cleaner look.

The iOS 26 style tab bar which I’d guess you may adopt later:

ios-26-music-app
ios-26-music-app1108×294 66.5 KB

Mostly at least for now I just hope for easier and other ways to view and search my history, as with Calendar.

1 Like
12

Hello Matt,

A few observations:

I come across places and activities preceded by an arrow, sometimes pointing up, sometimes pointing down and sometimes both. What does this mean?

I’ve noticed that after importing, I’ve had to reconfirm each venue and activity for practically every date for the last 7 years!

When in use, confirming locations and activities, my iPhone 15 pro heats up enormously and the battery melts visibly. I’ve tried relaunching the app and also a reboot of the iPhone but the problem persists.

I also had a crash while I was doing the confirmations but I couldn’t reproduce it. It only happened once.

Finally, as a big calendar user, I find the new calendar much easier to use.

I’ll be back if I find anything else a bit strange.

Siso

1 Like
13

Congrats on the launch of the public beta! Some initial observations:

Import took 27 minutes for me with 6 years of data.

The color scheme for paths in the map view in dark mode need work. The black line for car journeys seems very difficult to tell when the map is dark.

I also miss the gesture where tapping the map once causes the map to be full screen and the timeline view to be hidden; I really want that back.

When I switch between different days by swiping left/right on the timeline there is an annoying lag in the map view between re-zooming the map and redrawing the lines representing movement. The lines get redrawn quickly and then the map readjusts the zoom after an annoying pause. It’d be better to make it synchronized: so either both becomes faster or both becomes slower.

I also noticed that it sometimes doesn’t merge two identical stationary items with the same place. I’m not sure why. It also sometimes doesn’t merge two adjacent journeys with the same type of transport. For example why are there two Metro segments in this picture?

image
image1206×1202 46.3 KB

Also sometimes tapping the three-dots button in the timeline view doesn’t do anything.

After tapping on a photo in the timeline view, how do I get out of the full-screen photo view? I can’t seem to swipe to get out. Nor can I tap anywhere.

Good work with the ground-up rebuild though! Appreciate the extra debugging logs in the settings tab.

1 Like
14

Here’s potentially another issue: the underground portions of NYC subway is very punishing because of very weak GPS signal, and the old Arc app seems better behaved here. The new Arc Editor app keeps showing me a walking journey on the main timeline view that actually has two individual segments, one Metro and another actually walking:

image
image1206×2622 250 KB

There is however no way that I can tell to get the Metro segment shown on its own in the main timeline view. In the old app, one can go into the individual segments view and reconfirm the Metro segment to get it to show up but the new Arc Editor app doesn’t seem to work this way…

1 Like
15

Yeah, definitely. In Arc Editor I’ve added that second row to the Place search results when editing Visits / assigning Places to Visits. That’s a nice improvement over Arc Timeline’s single line results. Having the address below really helps in a lot of cases.

For actual Search view, once that exists I’ll also want it to have two rows of details for each result. And possibly more than that, depending on the kind of item found by the search.

Yep, there’s a lot more that could be done in the long press context menus. I also want to add that context menu to the “Timeline Cleanup” view. Sometimes I just want to quickly delete a Visit in the list there, but have to tap through to its edit view to delete it. It would be much quicker to just long press and choose Delete.

“Show in Maps” is interesting. You mean opening up that location in the Apple Maps (or Google Maps) apps? That makes a lot of sense.

Yeah the new iOS 26 style of having search button at bottom right looks good to me. I’m keen on doing that. Moving the Settings tab to … well, somewhere else, will also make sense I think. I’m holding off on doing any of that until iOS 26 is closer to release. But I’ll be wanting to add the Search view/feature quite soon, and certainly before iOS 26 goes final!

16

This is debug info that’s there pretty much only for my benefit. I’ll take it out of the beta builds soon maybe. It indicates whether there’s a “broken edge” in the timeline linked list. If the arrow is pointing up then it means that item doesn’t have any nextItem linked, so its “next edge” is broken. If arrow is point down, it means the same for previousItem being missing.

It’s basically just there to give me immediate feedback on whether the timeline linked list management is working consistently or not. With the processing engine doing all sorts of merges, edge stealing, etc, there’s lots of potential for the timeline to “break” in some ways. But LocoKit2 doesn’t have any known timeline linked list breaking bugs at the moment. So you should see those arrows disappear quickly, if they ever do appear.

That sounds very unusual! The import restores all of the confirmed ActivityTypes and Places, for Trips and Visits. So the only reason why you’d need to confirm any imported data is if the data in Arc Timeline app wasn’t confirmed. I suspect that’ll be what you’re seeing: data that wasn’t explicitly confirmed in Arc Timeline.

Immediately after the import finishes, the Place stats and ActivityType models will be missing, and pending rebuild. During that time the automatic assignments will be “low confidence”, thus being marked as needing confirmation. But once those stats / models rebuild the confidences of the auto assignments will go back up to what they were in Arc Timeline. So most of those “this needs confirmation” things will disappear.

Hm. When that happens, have a look on the Settings tab, in the “Active Operations” section. I suspect you’ll see a lot of operations listed in there, and some that are running too long (they should all ideally complete within only a few seconds).

I think there’s still some SQL optimisation work I have to do. LocoKit2 is much more efficient than old LocoKit, but it’s only had full multi year databases for a couple of weeks so far (since I finished the old LocoKit importer), so I haven’t had time to check every point of inefficiency yet. There’ll be a few I need to fix up still.

Hopefully it’ll show up in the Sentry crash reporting! That seems to be working quite well so far, so I’m getting good insight on any crashes. I think there’s two crashes at the moment, one common and one not, but both easy to fix. So hopefully I’ll have both fixed in the next TestFlight build.

Me too! That’s the iOS standard calendar view. Though frustratingly it’s very limited in functionality. I can’t annotate it with any information about the days. So it’s only useful for navigation and nothing else.

I’ll eventually still have to build a separate full screen calendar view like old Arc Timeline’s, for showing annotations on the days, to communicate extra information.

17

Agree. I think it’s mostly just the car colour, but some of the others could benefit from having special Dark Mode versions too.

Try swiping down from the top of the content sheet. That’s the iOS standard way of doing it - Arc Timeline was doing something non-standard, because I couldn’t get UIKit to work nicely with the swipe gesture at the time (many years back).

Arc Editor also has a new sheet position that Arc Timeline didn’t have. You can swipe the sheet up, to make it cover the full screen. That lets you view the timeline or details views without having the map taking up screen space.

For this one you’ll likely notice it works better some times, and other times not. What’s happening is the map update is sometimes getting delayed because there’s other work going on. Which means there’s still some inefficiencies in the system, that I need to iron out.

But yeah, also even when it’s running at proper speed there’s some timing tweaks it’ll be worth doing. I’m going to come back to the map updates at some stage and see if I can improve them a bit more.

Hmm. The two metro items… I wonder if there wasn’t enough data recorded, or the data was too low accuracy, so the processing engine is concluding “nope, those aren’t suitable for merging”. That’s actually something I’m keeping an eye on in my own testing: whether the processing engine is getting all the merges right, or at least getting them more right more often than old LocoKit did.

For when it leaves two adjacent Visits with same Place not merged, that one I’ve seen once or twice, and suspect it’s a straight up bug. But I haven’t caught it in a situation where I can debug it yet, so haven’t found the cause. It does appear to self resolve soon after in most cases though. But yeah, I suspect there’s some subtle edge case I’ve missed there.

Oh, on timeline tab that button always doesn’t do anything :joy: I haven’t got around to adding anything to that menu yet.

It’s supposed to be a swipe down gesture to close the photo viewer, but it often doesn’t seem to register. I’m probably going to completely rebuild the photo viewer. I tried to build it all in SwiftUI, but SwiftUI has some limitations that make it still not possible to build a properly functional photo viewer. So I think I’ll give up and end up wrapping an old UIKit approach, so that it can at least work properly, until SwiftUI is ready for the job!

18

Yeah I want to spend some more time testing underground train trips in Arc Editor. It’s a really difficult recording task, and needs a lot of careful testing and tweaking. I haven’t done that yet with Arc Editor / LocoKit2, so … in a lot of cases it’s making dumb decisions about how to handle the weak data accuracy / complete lack of data.

For splitting out segments like that, Arc Editor / LocoKit2 work identically to the old app. But I think the problem in that specific case will be that LocoKit2 didn’t record the data well enough, so it’s got a situation that can’t be cleaned up. Arc Timeline / old LocoKit probably would’ve got slightly better data, which would then allow the segment extraction to correctly lead to a separate metro timeline item.

So I think in that case once I do the testing and tweaking for underground train trips, the second problem should resolve as well. Well, I hope so!

I’ll be back in Bangkok in a couple of weeks, where there’s some good underground lines for doing testing on. I’ll see if I can make some progress on it then.

2 Likes
19

Terrific. As perhaps a fair amount may travel even to close cities merely from commuting or daily activities, perhaps such should be everywhere including timeline view?

I’m reminded there could be visits to the same place yet some from Foursquare and others from Google Places. Such visits are to the same place yet not and then total visits is incorrect. I’d guess you’re aware of such yet I’m not sure how to deal with it.

Like a contact or address book, perhaps some future view that is a list of all places visited? Where one can see place details, visit history and so forth as one can already and also perhaps mark some locations as grouped and have one as default? Not sure how good of an idea but I can’t think of anything else. Perhaps if some places have the name address and are close to each other, they can automatically be grouped in someway.

Agree. Reminds me and I forgot what was the thought before. Perhaps more so for locations that span a large area like a park or something, sometimes there could be a visit to a location, some walking, and then another visit, yet other times it just shows one visit. Perhaps its GPS accuracy that determines where it is one long visit or marked as multiple visits with walking (or sometimes incorrectly car) in between. From what I recall, maybe such walking before one could not delete and thus merge into one visit?

There is already such in Visit details I believe in the more menu. Titled something differently but if it seems (not sure) Show in Maps is more commonly used as text in other apps, maybe that. If so, maybe some option to change which maps app.

Some issues I noticed:

timelineItems.isEmpy on Today view then it refreshes. Once so far.

After initial import, I had an additional Home entry occuring right after. E.g., I was home for may be x hours (old data from import), then Arc Editor added a new Home visit right afterwards, without any walk stationary other in between. I was able to merge yet such occurred again the next day.

All in all, minor things and the new version is looking good.

1 Like
20

In this case you can use the “Place Merge” feature in Arc Timeline app (not yet added to Arc Editor). You can get to it from the Visit Details view, scroll down to Place Details, then from the ellipsis menu in there, “Merge places”.

That will find overlapping places and let you choose which one to merge that Place with. Which nicely cleans up the cases of there being Places from Foursquare, Mapbox, Google…

This behaviour is slightly different in Arc Editor. It’s slightly more keen to keep the Visits merged together for large indoor places (eg airports, malls). But for outdoor places like parks it’ll still favour having it split up into separate walks and Visits.

So that situation is slightly improved for large indoor places, but there’s still no solution yet for when you might want to merge several outdoor visits over a larger area (eg a park).

Yeah there’s a “get directions to” option. Oh and now that I think of it, there’s also a button to go to the Google Maps view for a Place, from the Place Details view, if the Place came from Google’s data.

Heh. I need to update that to have something more sensible, instead of that very debug text! Also yeah, sometimes there’s a case where the timeline data isn’t loading fast enough, so the list sits there empty even when there’s data that should be appearing. I think once I find the inefficiencies in some of the database queries that’ll self resolve.

1 Like