Hacker News Reader

Comment by VladimirGolovin | Hacker News Reader

Damn, now I need to move faster. My app must be ready by the time they close Wunderlist. I expected an eventual closure, but not that soon.

I've spent the last two years developing a todo app that would be a good replacement for Wunderlist (and Astrid, if anyone still remembers that).

The app will include with some major / critical features both those apps lacked: proper hierarchies / first-class subtasks, automatically-activated contexts (e.g. location), snooze for arbitrary durations, a robust implementation of recurring tasks, and an infinite calendar-like timeline to look ahead.

I went full waterfall on it, the spec is 266 pages long and took half a year just to write. I'm currently testing an Android alpha version, and, for me, it works ridiculously well, much better than Wunderlist, which sat on my homescreen since August 2013.

View on HN

> I've spent the last two years developing a todo app

Without seeing what you have, a general remark would be that everyone and his cousin has clocked at least some time developing at least one ToDo app. It's a right-of-passage thing for a lot of developers, just like a Weather app is the same for UX designer wannabies.

BTW, what set Wunderlist apart is not the features first and foremost, but their bold in-your-face marketing rooted in very pretty visuals and a fake arrogance routine. Whether they had proper hierarchies / first-class subtasks was quite secondary.

(OP here) I've set up a quick website with more detailed descriptions and some screenshots: http://tuskarr.tilda.ws/

I'm absolutely aware of the rite-of-passage thing, but I'm doing this because I NEED a proper, working todo app. My life will crumble without it. And I'm prepared spend my time and money on it. In any case, there's no going back to Wunderlist now.

As for marketing, I'm not worried. All I want is a working todo app, and it will be great if it turns out to be useful for other people.

I wish you had put an email sign up option in your site. You're missing out on potential alpha users.

Done. The signup box is up and running (scroll to the bottom of the page).

Add it to the top! People don't scroll that far down :)

Email form is there at the bottom.

Looks really nice. Unfortunately, I usually don't want my task list associated with dates. I just want a list of stuff I need to get done (actually, A lot of lists of stuff). Once in a while I'll want to associate a reminder date but that's about it for the way I work.

Have you tried WorkFlowy? It is basically lists of lists.. of lists and so on.

Never liked there UX though, but the concept.

It looks like something I would use, so I hope you're able to get it out there.

My 2 cents - "Tuskarr supports first-class subtasks" might be a fine description on HN, but an average user won't know what that implies

I wrote this text in an hour, specifically for HN :) Of course the description for the general public will be more accessible - and supported by screenshots.

(Which reminds me that I should be doing the screenshots now.)

The combination of the screenshots and sales pitch convinced me!

This sounds awesome, I'm looking forward to testing!

Break a leg.

It is fine, occasionally a neat idea or concept emerges, but I just use Org mode and Agenda now. I always see or hear people eventually giving up on Emacs for some pretty new editor (VS Code) or finally quitting vim because they "feel more productive" somewhere else. Or using some new and random TODO app. Meanwhile, I have been using the same system for 7 years and I know it will work on any platform any time and my data is all plain text and not under control of any third party. Finally, I have a depth of features most TODO apps /still/ lack. But there is a huge reluctance to embrace this work style, but I think it is a short term gains / long term gains problem and most people end up with whatever is shiny and in the moment, but it will get abandoned over the long haul. My text based system will work in 10 years. Will whatever shiny and new TODO app?

You're right of course. I guess I'm going to give Orgzly another try.

> what set Wunderlist apart

A nice wood texture (not even joking, was totally sold on the visuals in the early days)

> just like a Weather app is the same for UX designer wannabies

20 year UX veteran, this is the first I've heard of this (o; Not a criticism, certainly made me smile, but I've not heard or seen this before.

(that said, I work mostly in enterprise, this might just be a mobile thing)

Behold... https://dribbble.com/search?q=weather+app

I'd argue that's visual design. A subset of UX.

UX primarily starts with user research, then a metric shiton of postit notes, paper prototypes, a bit of testing and then some hi fidelity mockups

> Without seeing what you have, a general remark would be that everyone and his cousin has clocked at least some time developing at least one ToDo app. It's a right-of-passage thing for a lot of developers, just like a Weather app is the same for UX designer wannabies.

Its funny, because it true. :D

"rite of passage"

right of passage is probably more like the right of way :)

Looks like it switched in the 1970s

https://books.google.com/ngrams/graph?content=rite+of+passag...

I'm no language expert, but I don't think there was any "switch".

As Wikipedia says, "[rite of passage] is the Anglicisation of rite de passage, a French term innovated by the ethnographer Arnold van Gennep in his work Les rites de passage."

"Right of passage" may mean something similar to "right of way", and may be used my mistake when people mean "rite of passage" but there was never a switch in how "rite" is spelled.

No, it didn't switch. "Right of passage" is a thing, but it's a completely different thing from "rite of passage".

Compare:

https://en.m.wikipedia.org/wiki/Right_of_passage

https://en.m.wikipedia.org/wiki/Rite_of_passage

Indeed. Sorry for a brain fart.

What set Wunderlist apart is that they have been the archive of my project ideas for the last who knows how many years. While other apps came along and disappeared, I was sure Wunderlist was just "there" when I needed it.

>I went full waterfall on it, the spec is 266 pages long and took half a year just to write.

Wow. It's nice to see extensive planning like that. I usually just throw anything that seems like an important idea into a text file on the assumption that it'll be fully hashed out during production at a later time. That method entails far less work, with the caveat of occasionally wondering if you're delusional from time to time.

In any case, good luck. I hope you kick ass. If Microsoft employs their usual design aesthetic, doing so should be very possible.

In case of this project (btw, here's a temporary website: http://tuskarr.tilda.ws/), it was impossible to quickly throw it together. I had to think it through in one single chunk.

The app combines four intersecting concepts:

1. Infinite timeline with addressable days. 2. Recurring tasks that support this infinite timeline. 3. First-class subtasks, which have all properties of normal tasks. 4. Contexts, which hide tasks from the list based on certain conditions.

I simply couldn't create a working mental model that accounts for all possible interplay of these concepts without speccing it out in advance.

IMO your biggest competition is probably just whatever comes installed on the phone.

Closely followed by the top 5 "todo" apps in the app store with four star ratings or higher.

> the spec is 266 pages long and took half a year just to write

How many times did you make paper/cardboard mockups and user-test the interactions?

Quite a lot. Instead of paper, I preferred on-screen prototypes created in Photoshop and displayed as images on a phone screen.

Check out: http://www.appcooker.com

One thing that is great about wunderlist is the truly instant sync. Have you been able to replicate that? I vaguely remember reading an interview somewhere about how they accomplished it.

Except that it often does not work, when working on shared lists.

Yes. We're using Realm for realtime sync.

I'm wondering if you're including the Astrid feature I still miss and have never found a replacement for: assigning locations to tags, so that you can do things like have an alert come up with "stuff I gotta do at home" when you get home.

If you have this and an iOS version, then you have my money.

It's too late to edit my original comment, but I've set up a temporary website, where I explain this app in more detail, and show some (temporary and very messy) screenshots and designs:

http://tuskarr.tilda.ws/

Put a signup box for an email list on there, so that we can be reminded once it's closer to release

Doing this right now, should be up soon. UPDATE: The signup box is up and running.

This does look interesting

And then one day a tech giant will contact you for acquisition and ....... :|

Yes, I'm aware of this problem. First, we're unlikely yo be acqui-hired, because we're based in Russia. Second, I simply don't believe that this app will reach the levels of success similar to Wunderlist or Todoist :)))

But I definitely think that there should be a legal way of protecting existing users of an app from "Our Incredible Journey". Maybe some clause that permits the authors to open-source the app.

Are there any examples of this in the industry?

Isn't that what happened with Astrid? It was at least part open source and they released that code when yahoo acquired it.

Of course, the buyer may just ask you to remove that clause prior the acquisition.

Your barrier in gaining popularity once you launch is going to be cross-platform support. It sounds like you started wholly with Android, with other platforms coming further down the pipeline. These days a product that does not have apps for at least Windows, Android, iOS, and OS X severely limits its customer base.

Android essentially represents only a subset of phones and tablets. An app like yours is useless to many people unless it has desktop equivalents, as mobile-only access heavily restricts a user's ability to be productive where and when they need it. I might be an edge case, but I use Windows, OS X, an iPhone, and an Android tablet. Your app could be the most functional and usable todo app out there, but a competing product with 10% of your features is going to wind up being my choice simply because I get to use it across all my workflows. Platform support beats out the quality of the app, every time.

Based on your screenshots, the app is wonderful. I simply suspect you have focused too much on launching a 100% complete product instead of starting with an MVP. Obsessing over having a perfect product that launches too late to market, while your competition launches early and takes your cake, is all too common a scenario.

Yes, we started on Android, because that's the most efficient way for me to dogfood the app. Another reason why we didn't start developing for more platforms in parallel is that the concept was unproven. I wasn't sure it would work.

(Under "the concept" I mostly mean the approach where all non-snoozed or dateless tasks are shown in Today, and you swipe away the tasks that are not yet "ripe" to be done -- whereas in traditional apps you do the reverse - you select tasks to do by starring them during a review).

Now that I found out that the concept clearly works well at least for one user (me), the next step would be to validate the concept on a wider group of users. An Android-only soft-launch, or even a closed beta should be enough for that.

And if the concept is valid, we'll definitely develop for other platforms, the ones you mentioned. iOS and Android clients will be native, and the desktop apps (Windows and macOS) will probably use some cross-platform technology.

And thank you for the kind words :)

First of all, clearly the advice I'm giving is a bit too early for where you are at presently. I think that's giving credit to you for the quality of the work you are producing. Basically, I want this app. You might want to dial back just a tad on the intricate details, out of fear you may wind up never shipping an initial version. Being a software dev myself, I understand that you could spend another year (hell, even 2-3 years) completely hashing out every minute detail. However I think that what you have is getting to the point of being "above and beyond what is necessary" that you need to stop tinkering so much, and find a way to ship something sooner rather than later.

When you get around to desktop clients, ignore the seemingly negative consensus regarding Electron[1] if you think it's the right fit for you. All the negativity I've read about it is from "purists" who aren't specifically singling out Electron; rather they are hating on all non-native implementations. I noticed the mention of Xamarin on your site; never heard of it, but I'd be weary as they seem to be oriented more towards mobile whereas you're looking for a cross-platform desktop solution which is where Electron shines.

If your goal were to be the gold standard for apps, you should be developing using native OS X APIs; the performance you get from a native implementation on OS X is astounding. However, if you do native OS X you're putting yourself in a position to do native Windows and possibly Linux apps, and the collective work required for native-everywhere obviously does not come easy or cheap - particularly if you are doing this as a solo project without assistance from a team.

Electron is a perfectly fine balance for most UIs that are fairly simplistic, which is going to be true even for the most advanced todo-type app. Extra bonus being that if you put in the effort to plan out the Electron app side-by-side with your web/browser-based app, you can share the frontend development between the two, resulting in less overall work.

[1] https://electron.atom.io/

Being a software dev myself, I understand that you could spend another year (hell, even 2-3 years) completely hashing out every minute detail.

I won't do that :) I stopped working on the spec the moment I handed it down to coders, and my total time spent on the spec since then is around one week. When the spec is written well, and you simulated all important situations in your head before encountering them in the wild, you actually don't need much extra time.

There's almost no tinkering happens on the app right now. I'm fixing critical bugs and smoothing out rough details that manifested after months of my dogfooding.

Regarding the desktop solutions: yes, going native would be perfect, but that is not feasible for a 4-man team. We'll go with Electron or Xamarin or, less likely, Qt - we haven't decided on the exact technology yet.

Sounds great! I hope it will succeed. The MS cloud is something I prefer to stay away from.

I like your list of features. I currently use OpenTasks on Android, because it's the only thing that syncs (via DAVdroid) with Nextcloud. But while Nextcloud supports task hierarchies/subtasks, OpenTasks doesn't. Will it be possible to do DAV syncing with your app?

[For heavy-duty use and project planning, I use org-mode. But there's still no really satisfactory mobile org-mode story.]

If you need another tester I'd be happy to help out :)

Currently I use OmniFocus, primarily due to the start dates which hide things until they're relevant, which so many apps (including Wunderlist) lacked. Sadly the Android support is poor and only covered by 3rd party clients that lag massively behind in functionality.

Thank you :) I've added a beta signup form at the bottom of the page: http://tuskarr.tilda.ws/

> I went full waterfall on it, the spec is 266 pages long and took half a year just to write.

Why would you write a ~9k “loc” long spec over 6 month instead of coding it directly ? The only reason I can think of, is if you want to out-source the development. Is there any good reason to work this way on a one-man project ?

Not OP, but once you have a spec, it's much easier to code up. If you're designing while you're coding, you have to switch all the time between 'how should this work' and 'how do I build this'. Thinking and writing down how something should work, should always be the first step - regardless of whether you're on your own or in a big team.

(before the obvious objection: exceptions are toy programs which you write to find out how to build something, in which you focus on the technical aspects. The technical aspects ('how does it work') don't matter for the end user (= your potential customer), they only care about 'what does it do'.)

It was a hard lesson to learn, but "Weeks of programming can save you hours of planning" holds true in almost all cases.

In my experience there is far too little respect for well written specifications.

Most colleagues of mine and a few managers see them as a waste of time ("you're over thinking this", or "we'll figure it out later") yet somehow the projects with a specification always end up being done more or less on schedule.

The thing about specifications is that you need to be open to changing them if you find it's not working out. Specifications are not synonymous with waterfall development, and it's harmful to treat and view them as such.

I totally agree with this statement, but here we're talking about 0.5 year of planning !

Planning is important, and cautiously thinking about your features before diving in can save you weeks later on. But spending 6 month in a tunnel, writing specs you'll spend at least a year to implement, is a huge risk : what if your users only use 30% of the features you designed, and lack a missing feature you didn't plan ? You could have figured it out way sooner have you built something earlier and released it.

Also, your time to market is way slower and you don't risk losing years of work by missing a deadline …

> Damn, now I need to move faster. My app must be ready by the time they close Wunderlist. I expected an eventual closure, but not that soon.

> what if your users only use 30% of the features you designed

Or worse, what if different segments of users only use 30% of different feature sets? See also: Evernote's 5% problem [1]

[1] https://venturebeat.com/2016/01/05/evernotes-5-problem-offer...

> I totally agree with this statement, but here we're talking about 0.5 year of planning !

Yeah but that's not necessarily in man-years. Maybe he worked on it an hour or two a day. Now we're talking around two man-weeks of planning.

Well you'll only find out which 30% after building it, regardless of whether you have a spec or not. You can write a.spec for an mvp too.

I never plan. I just code. I end up writing 10 versions of every module, and retrofitting everything that touches it 10 times as well. What I'm trying to say is that I think... I think you might be on to something :)

> Is there any good reason to work this way on a one-man project ?

(Not the OP.) At least for me, emphatically yes.

From your question, it sounds like you're only considering the communicative value of a spec. That's of course important, but I consider it secondary.

Writing a good spec forces you to think through what you're developing in an orthogonal way. You're thinking through flow, cohesiveness, design, some implementation, UI, etc. together at a high level.

I love having the time to write a spec when I'm building something nontrivial. I've found numerous problems that I wouldn't have otherwise by doing so, ranging from simple UI problems/contradictions to spotting unconsidered assumptions that would have lead me down dead ends.

A spec and the application should naturally co-evolve during development to some extent. But the spec also provides an anchor for what you're doing, as well as a map. It is very, very helpful if you Do It Right(tm), no matter the size of your team.

Writing a good spec forces you to think through what you're developing in an orthogonal way.

(OP here). This. A thousand times this.

Did you follow any standards when writing? I am currently looking for ways to document and model our process and business rules.

I just wrote well-structured, dense, coherent / "mentally type-checked", bulleted-list prose, often copied directly from Workflowy. Some of the critical parts were described with FP-style pseudocode, for greater exactness.

thanks!

It's so you won't have multiple 'ooh' moments a year in coding - "I need to change that base thing, will require large rewrite". Or at least decrease possibility of it. Six months seems a bit too long even for 266 pages specs though.

(OP here) This is not a one-man project. I'm the founder, and there's a UI designer and two coders - which makes it 4 people total. And we'll add another dedicated coder for iOS development, when we find a good one.

First item on Microsoft's "To-Do" list:

- Shut down Wunderlist

Now let's hope they procrastinate just like everyone else.

I think the point of Wunderlist was the online synching across devices and the list sharing. It was that there was some trust built up with the brand. I wouldn't just start using some random todo app and allow it to sync my todo lists without having some level of trust built up.

Something else that was critical is that I could use Cortana to add tasks to lists while driving or without typing on the keyboard.

I think when I want to go deeper in organizing tasks, todo apps are often bad for that simply because a document format tends to do better once you're in the thick. It's better if those things are stored, tracked and organized with the work itself most of the time and todo apps tend to be outside that context.

Nice looking app, but screenshots are not much use. Do you have release anxiety? 2 years and nothings out but a landing page?

If you had released it earlier you might have been able to get some traction right about now when people are forced to leave wunderlist.

You want to move faster, then release the app. Customer feedback positive and negative will get you making changes quick smart and you'll get a good sense of initial demand. App development is a continuous process that requires constant input from users. No point in adding features no one asked for, you're just wasting your time.

As a wise man once said 'It's better to have released and failed than to never have released at all'.

wunderlist was the only todo app for me that stuck, mostly because it was simple to use, i would drop allmost anything to wunderlist without much structure. and thats why i kept using it. keep the advanced feadures but dont let them cluter the UI .

Would you be willing to share the alpha? I've been a long time fan of wunderlist and just put a ton of things into it for an upcoming move. Kind of bummed it's being shut down and would love to try yours out.

Yes, of course. Here's the beta signup form, at the bottom of the page: http://tuskarr.tilda.ws/

Hope your waterfall spec has caldev integration. I want a to do app that syncs with my iPhone todos and Google todos. I should be able to tell Siri to add something and mark it off later from my desktop. Or add couple of things with Siri, some auto populated from Gmail and notes.

Basically I don't give a shit about a walled garden to do app anymore no matter how good it is. I don't want a new app. I'd rather use something that puts my existing apps on steroids.

I wish you get lots of users in your app.

And somewhere in several years it'll grow big.

And Microsoft or Apple will buy it.

And someone will need to move faster to release his to-do, which is much better, of course.

I do remember Astrid! it was my goto app when I first got a smartphone.

I use WP8 day to day but dabble in Android, I'd love to give your app a spin...

astrid is open source now and available on play store.

Does HN have a way of moderating astroturfing? I'm honestly curious. Im aware of the typical HN digressions but it appears lately, threads are just consistently hijacked encircled in twaddles and responses to channel traffic and discussion off to some app.

What are the checks and balances if any to keep the astroturfing binge to a minimum and encourage discussions instead of traffic channeling bait.

Yes, there is moderation. If you have concerns about astroturfing, you can contact the mods via the Contact link in the footer.

#!/bin/sh\n

vim ~/todo.md

Reply on news.ycombinator.com