A roast of Colony: why it failed and what to do next?

Development
1

Now that IF has ended, I am taking a step back to assess my current situation. Outside of the IF reward, Colony has been an abject failure in my opinion. The hope was that people would upload their content and attach metadata, but it seems that for the most part people prefer to upload data and share addresses manually on here and Discord. Only one person uploaded a pod and shared their address with me. I am aware of no one else that is using the application today.

All this is fine. I want to make something that people find valuable and will use. The Colony GUI app for me is mostly a test vehicle and learning exercise. In that regard, it was a great teaching tool, I learned a lot. The library under the hood that powers the metadata sharing and the indexing of what exists on the network today is what matters to me. These pieces can be plugged in elsewhere.

So to anyone that tried the Colony app, what did you like? What did you hate? What was clunky? Any GUI design ideas I could use (I am not a UI designer, any help here is GREATLY appreciated :smile:)? What features are missing? What features werenā€™t valuable? Or maybe it wasnā€™t useful because there wasnā€™t enough stuff to search. Does it just need more content?

Honestly donā€™t hold back, anything you can provide is useful. I need help determining if I should continue developing this app as is, starting UI development over, repackage colonylib into some other application, or bail on this idea entirely and work on something else for the network.

12 Likes
2

oh man, I would not abandon the project, definitely not.. I tried Colony, it was hard for me to understand and make things uploaded to the network - many attempts with many failures. Even with this new Dave release, I am not able to upload. So I just decided to wait and try all these apps later. I found your Colony to be a great idea but do not be disappointed with small engagement - to be honest, there are like 50 active people on this forum, you canā€™t expect miracles yet.

For me what was hard to understand - was the whole idea of pods, what it is, what it does, what should I do, etc. You are a pioneer, and I am just end user waiting for things to work nicely in GUI, waiting for less expensive uploads, etc. That is my reason why I stopped engaging for a while, but I definitely have intentions to try using all these apps again soon.

10 Likes
3

Thanks for the feedback. I really struggled with how to represent the idea of a ā€˜podā€™. I honestly hate how the UI turned out in this regard. I wanted some kind of folder/file tree thing, but couldnā€™t figure out how to represent this in the UI at the time. My UI guy that did the frontend said we didnā€™t have time and so we slapped together what we had and ran with it.

If I were to do it again, I would change ā€˜podā€™ to ā€˜folderā€™ or ā€˜directoryā€™ and a ā€˜pod referenceā€™ to a ā€˜linkā€™ or something similar. At a high level thatā€™s kind of what these things are doing.

Plugging in the metadata was tedious and weird too. I never figured out how to attach the metadata upfront when a user uploads, but that would be the ideal case. Even better if this was automated somehow.

6 Likes
4

Iā€™m sorry to admit I didnā€™t try Colony, or any of the other IF apps. I could see that most werenā€™t ready through the process and didnā€™t have free time enough after they were deployed. So I cannot judge Colony or any of the others, winners or not.

I can still say that it is far too early to judge any of these apps in practice because there are too few people available to try them out.

Iā€™ve been very surprised how few people have published websites with the dweb-cli, which has been possible for about a year. But it is not a failure - fight me!

I think the idea behind Colony is sound, and that all of us need to work on ease of adoption ready for the arrival of curious normies, or they will stay not being used.

Thatā€™s my focus and I have a new app ready to publish that will take us, especially dweb apps a step closer to everyone which is my top priority.

We have a good set of early offerings, but my worry with the IF approach is that it will generate potentially great apps that do not continue to be developed. Which are dropped because the motivation behind them was to win IF and not to build something sustainable and continue developing and promoting it. The latter is a large part of the work by the way.

That IMO is what Autonomi have yet to realise, but I get told to shut up for pointing it out, as if a few tweaks to IF will change this fundamental problem.

We donā€™t just need good apps, we need a way to get builders who want to build forward, promote them and provide ongoing support, so they can sustain them way beyond a short competition period.

Weā€™ve seen how slow ā€œlaunchā€ is now, and this will continue IMO. We cannot expect a big bang, so we must plan for slow build up.

So we need programmes that support developers over time. We had the Bamboo Garden fund donated explicitly for that, but I believe it has been spent on the core. Earlier there was a plan to set up developer Pods. Both bang on IMO, but now we have short term competitions and no idea if or how these will be followed up effectively.

We could use IF to identify projects and developers who merit support. I think we should not just keep running IF competitions and then drop the winners and give them no incentive or support to continue.

14 Likes
5

Guys, Iā€™m not a programmer so forgive me if Iā€™m being ignorant but there is so many great talents and programmers here - canā€™t you just all come together and develop one big beautiful opensource app, that you will make just perfect for us normies?

Letā€™s say everyone may continue developing their own projects, like @happybeing with dweb and @zettawatt with Colony. But what about running a referendum about what the most important app could possibly be for Autonomi, call it a killer app, and then you geniuses joining all together to make the app just perfect, with awesome UX and GUI?

Could there be a consensus like this or am I just naive?

11 Likes
6

Thatā€™s my vision for the Autonomi App Co-Op. Weā€™ll see if others agree. My hope is that we can combine the pieces weā€™ve made into something epic. Weā€™re all using the same technology stacks, so it isnā€™t a technical problem, its an organizational and vision issue. What do we want to make and how do we go about it?

That is a key motivator for this topic. I want to roll Colony into something better that people will use, but I need more insight from users about what did/didnā€™t work so it can be better.

Edit: to be clear, not just an app, a platform that makes it easier for everyone to build on.

10 Likes
7

The only way to do that will I think be centralised control. Also, thereā€™s no perfect app that will achieve your goal. We will always need an ecosystem of apps because users are different and will want different things.

We will also need to support failures as well as successes, because we wonā€™t know which will be which when we support them.

Developer Pods were a great way to help build a sustainable ongoing ecosystem IMO.

2 Likes
8

@zettawatt I tried all IF apps and colony was one of my favorites.

I dont think that you should let use or lack thereof get you down at this point, it is certainly not a failure.

I think the issue is that our community is small and out of that small community a large percentage have their own projects going on so time to use other apps is tight.

My opinion is that this is a good problem to have. Keep grinding, Colony is a great idea.

13 Likes
9

I feel we need to grow the community.

There are a lot of people but many are from years ago that may have not kept up with the tech and news.

The benefit of using a crypto token is you get to use their ecosystem for marketing. However for some reason arbitrum was chosen, which has a tiny ecosystem. If we had went for Solana or something, then we could have marketed in places with tens of thousands of people and huge amounts of capital for investment, which would have brought a lot more eyes.

5 Likes
10

So to be clear, your solution to fund app developement is to have a central entity send payments to app developers? Iā€™m not trying to be confrontational here, Iā€™ve heard you say this before, I just donā€™t understand the logistics of how this is supposed to work or where the money would come from.

11

My recommendation is to make a clearnet Coluny site portal through which you can upload and access content. This idea that there will be enough people using the network directly in the coming years has a high probability of failure. It is better to be safe and bet on a hybrid solution.

Check out the Impossible Futures!

5 Likes
12

Please do not give up on Colony.

I probably try more new apps/projects than most and frankly I do not have time to give them all the attention they deserve.
That was/is certainly the case for Colony.

Some more advice/help on setting up Pods may help as well.

Keep at it , please.

6 Likes
13

Going to chime in here

Iā€™ve not (yet) been able to find the time to test Colony yet, Zettawatt. But from what Iā€™ve seen youā€™re (and so many others) one hell of a developer. I think the problem in general (not just colony) is that weā€™re missing some good UX/Product people in this community. Developers (And I mean this in the best possible way so donā€™t get me wrong) usually lean extremely tech-heavy and what is intuitive for them makes absolutely no sense for the outside world.

I saw your CTA for the developer co-op and Iā€™ve been thinking hard if I can commit to contributing with my product expertise. Honestly, Iā€™ve not decided just yet, but I feel it would greatly help user adoption in so many ways that I almost feel obligated to make time for it. And again, donā€™t get me wrong, Iā€™m mediocre at best at what I do, but the expertise just seems lacking here overall.

You can hit me up any time at Discord (or here) if you like and let me know what you might need. Iā€™ll at the very least try to guide you into the right directions. That pretty much counts for all developers trying to build something amazing on the network, Iā€™ll help if I can. If youā€™re targeting end users (and not just developers) itā€™s good to have a general guideline: ā€œif Iā€™d show this to my mother or son, would they be able to understand what Iā€™m building?ā€ The answer should be yes. always.

9 Likes
14

Yes I agree with Dimitar. I personally would like to backup my videos on Autonomi, then share a link with my friends, on which they can easily play the video in browser.

But so far I struggle with uploading and definitely I struggle with creating such a link. I tried Dave but I was not able to upload anything yet. With Colony, I was able to upload after some time, but I have no idea of how to share the link to my video in easy way, so that my girlfriend can play the video in her browser.

2 Likes
15

Yes, this is spot on. I am not a ā€˜normalā€™ user and I really struggle with UI/UX. To be honest, anytime I interact with the colony datasets, I typically just hop onto the CLI version that I made, that is my preferred UI :rofl: If someone was to draw up a good clean interface, I could build to that, but I would never be able to come up with a good one myself. I think thatā€™s why Iā€™m somewhat disgusted with Colony at the moment. It just doesnā€™t look good to me, I donā€™t know how to fix it, and it is extremely frustrating.

I like this idea a lot. I have the backend colonyd that has REST endpoints for all of the functions. Iā€™ll think on this more.

2 Likes
16

I dont think iā€™ts inertia, rather a case of getting too many presents all at once, and everyone needing a bit of time to adjust to the pace of change and development now happening, so everything is going into the basket (Davids words) and we will see what gets made from the ingredients is how i feel, and if i havenā€™t sampled your ingredients yet itā€™s not because i dont want to itā€™s because my shoppng trolley is about to tip over haha

5 Likes
17

Perhaps just clone an existing interface that users are used to?

Dropbox or something alone those lines maybe, clean minimalist. As you mentioned before get rid of the terminology for generic terms like files, folders and information. Make it so metadata is automatically detected (we did this in jams, not sure if you have that working).

Have the search just be like a normal folder view search input. When you load a pod (just have it as a folder full of content and the search will look through that folder, or the entire computer (all pods).

For the OS I am planning to make it so the users donā€™t even know they are using Autonomi. Just make it like a normal desktop/folder experience as all the big companies have spent decades getting the population used to that already.

12 Likes
18

Iā€™ve only tried a few apps like anttp, Jams and tried to install another one but got stuck with too complicated steps at that time.

So I thought to wait until after IF was done and hoped that apps would be easier to install, better interface and that the network performed better about finding files, not failing on download and such.

The first thing I want to try are the upload/download apps and then try the other ones. Backup of important files is the first thing I want to do on Autonomi. I have not had the need to upload any files yet with meta-data but want to give it a try.

For people that are more serious about structured sharing then Meta data is more important.

3 Likes
19

I donā€™t think thatā€™s confrontational, but it is a bit straw man.

There are always flavours of governance, many ways to gather and share funds that are not quite as authoritarian as that :laughing:

Iā€™m not getting into ideology. My point stands, that what we have is a central entity (irrelevant to me at this early stage - just as for the core development) using funds in a way that doesnā€™t look like delivering what is IMO needed.

And what we need at this stage, IMO is a way, to provide not just shots in the arm that quickly wear off, but ongoing support until the ecosystem itself is big enough to attract and support users and developers. After that it becomes decentralised because the support is no longer needed or effective.

5 Likes
20

Got it. So basically MaidSafe has funds now, we have projects being developed, funds should go to support development until the network can stand on its own 2 legs. Makes sense.

2 Likes